Helix equations

Hans Drevermann

April 23, 2007

1 Abstract
The representation of helices in cylindrical coordinates can be very useful for the graphical
representation of helices and for other applications.
We will show the graphical representations of helices in cylindrical coordinates, i.e. the
projections φ/ρ, φ/z and ρ/z.
We will give exact and approximate mathmatical representations of helices in carthesian
and cylindrical coordinates. This leads to a better understanding of the respective graphical
representations.

2 Introduction
Charged particles move in a magnetic field along curved tracks. Assuming an ideal ho-
mogeneous field and neglecting multiple scattering the tracks are described by helices.
Depending on their charge the particles run clockwise (in ATLAS negative charge) or
anticlockwise (in ATLAS positive charge) along their helices.
In ATLAS, as in many other experiments, most particles, mainly the interesting ones, leave
the magnetic field after being only slightly bent. Thus, tracks are just helix segments.
Practically all tracks of interest from one interaction in high energy physics stem from the
same primary vertex or have an origin close to it.

3 Circle equations in carthesian coordinates

We define the z-axis of a carthesian coordinate system to run along the beam line, i.e.
parallel to the magnetic field lines. We assume that the primary vertex lies exactly on the
z-axis, i.e. it has the coordinates x = 0, y = 0 and z = zv .
Figure 1 shows the y/x projection of a helix which is a circle1 . The circle passes at a
minimum distance D0 from from the center of the coordinate system at P0 . A particle
running along this circle starts in the direction φ0 . Therefore we give also the circle a
direction φ0 . A circle with a direction can be anticlockwise as in figure 1 or clockwise.
We will first discuss a circle which starts exactly from the primary vertex, i.e D0 = 0, so
that the points Pd and P0 fall together. For an anticlockwise turning circle the angle ∆α
increases and ∆α > 0. We call this a positive turning circle with sign S = +1. Attention:
this represents in ATLAS the track of a negative charged particle. A negative turning circle
with decreasing azimuthal angle and ∆α < 0 has S = −1. For a positive (anticlockwise)
turning circle the direction of the vertex as seen from the center of the coordinate system
1
Figures are found in black and white at the end of the paper.

1
 π π
P0 is φ0 + . For a negative (clockwise) turning circle the direction of the vertex is φ0 − .
2 2
R is the radius of the circle. The center point Pc of the circle has the coordinates xc and
yc (see figure 1). xc and yc are given by:
π π
xc = R · cos(φ0 + S · ) yc = R · sin(φ0 + S · ) (1)
2 2
xc = −S · R · sin φ0 yc = S · R · cos φ0 (2)
With R0 = S · R one gets:

xc = −R0 · sin φ0 yc = R0 · cos φ0 (3)

If the circle does not pass through P0 but in a distance of nearest approach D0 (see figure 1)
equation 3 changes to:

xc = −(R0 + D0 ) · sin φ0 yc = (R0 + D0 ) · cos φ0 (4)

This can be directly deduced from figure 1. If R0 and D0 have the same sign P0 lies outside
of the circle, otherwise inside. The coordinates xd and yd of the point of nearest approach
Pd are given by:

xd = −D0 · sin φ0 yd = +D0 · cos φ0 (5)

The coordinates x, y of points on the circle with an angle ∆α as seen from Pc are given
by:

x = xc + R0 · sin(φ0 + ∆α) y = yc − R0 · cos(φ0 + ∆α) (6)

As a test, one gets from equation 4 for ∆α = 0 the coordinates of Pd given by equation 5.
π π
To draw a circle segment of from Pd for positive turning circles ∆α varies from 0 to + ,
2 2
π
for negative turning circles from 0 to − . R0 and ∆α have always the same sign.
2
In HEP experiments one may find tracks of very high momentum, so that R0 approaches
infinity. Measuring errors of the space points, from wheach the track was reconstructed,
may increase R0 furthermore. In this case the use of the equations 6 may lead to errors,
as one has to deal with small values being the differenc of very large ones. This is avoided,
if one calculates points on the helix by use of equations in polar coordinates.

3.1 Helix equation in carthesian coordinates

Giving the points of the circle a third coordinate z one gets a helix. z is defined by:

z = z0 + tan λ0 · R0 · ∆α z − z0 = ∆z (7)

2
where z0 is the z coordinate of Pd and λq
0 the angle between a tangential line of the helix
PZ
and the xy-plane ( tan λ0 = , PT = PX2 + PY2 ). As R0 and ∆α have always the same
PT
sign, ∆z and tan λ0 have the same sign.
In a homogeneous solenoidal field R0 is proportional to the strength and direction of the
magnetic field B (negative in the case of ATLAS) and to the transverse momentum PT of
the particle.
PT
R0 = (8)
s·e·B
where e is the elementary charge and s the sign of the charge of the particle. Then one
gets:
PZ PZ
tan λ0 · R0 = · R0 = (9)
PT s·e·B

Attention: Pd is not the point of nearest approach of the helix to the center of the coordinate
sytem, i.e. in 3D, but the point of nearest approach of the circle to the z-axis, i.e in the y/x
projection, which means in 2D. This 2D point becomes a 3D point with z0 as z-coordinate.
A helix, which has at the point of nearest approach Pd the coordinates xd , yd ( see equation
4 and z0 is defined by the paramters R0 , D0 , φ0 , tan λ0 and z0 . They are called perigee
parameters. Instead of R0 sometimes R10 or s*PT are used. Attention: there may be
different definitions of the signs of R0 and D0 . Instead of tan λ0 one may use a function
of tan λ0 as e.g. η. This must be taken into account, when applying these equations. The
perigee parameters describe a helix, but not a helix segment with end points.
In HEP experiments one may find tracks of very high momentum, so that R0 approaches
infinity. Measuring errors of the space points, from which the track was reconstructed,
may increase R0 furthermore. In this case the use of the equations 6 may lead to errors, as
one has to deal with small values being the difference of very large ones. This is avoided,
if one calculates points on the helix by use of equations in polar coordinates.

4 Helix equations in cylindrical coordinates

A point P on the circle has the polar coordinates ∆φ’ and ρ’ as seen from Pd and ∆φ and
ρ as seen from P0 (see ed h2.pdf a):

tan φ = y/x ρ2 = x 2 + y 2 (10)

The φ value of a point on the circle starting at Pd in a direction φ0 is given by (see figure 2b):

φ = φ0 + ∆φ (11)

3
Figure 2b shows φ as a function of ρ in red, if D0 = 0, and in white, if D0 6= 0. The white
line superimposes the red one for larger ρ values. The green line is the linear extrapolation
of the red one, in order to illustrate the non linear behaviour of the white and the red
curves for larger ρ values.
ρ and φ together with the third coordinate z as defined in equation 7 define the cylindrical
coordinates of points along the helix.
To show a function v=v(h) graphically, a graph of v on the vertical axis versus h on the
horizontal axis is normally presented. If one shows points, whose position is described by
the function, one speaks of a v/h projection of the points and in the same way one can
speak of a v/h projection of the function.
It is of interest to understand φ as a function of ρ: φ(ρ), as well as φ as a function of z:
φ(z), and ρ as a function of z: ρ(z).
Figure 3 shows a helix in 3 projections: φ/ρ (figure 3a), φ/z (figure 3b) and ρ/z (figure 3c).
The respective magnified projections are shown in the figure 3d to figure 3f.
Three regions are clearly visible:

• the inner region, where

– φ/ρ and φ/z deviate strongly (by 90o ) from linearity,

– ρ/z deviates very slightly from linearity;

• the middle region, where the functions are linear;

• the outer region, where

– φ/ρ and ρ/z deviate from linearity,

– φ/z remains linear.

In the following the equations for φ(ρ), φ(z) and ρ(z) will be given

• exactly: covering the three regions

• approximately: for each region separatly.

4.1 Exact helix equations in cylindrical coordinates

A positive turning circle is shown in figure 2a. From the figure one can derive 3 basic
equations:
∆α
∆φ0 = (12)
2

4
 ρ0
= sin(∆φ0 ) (13)
2R0
With equation 7 (∆z = tan λ0 · R0 · ∆α) one gets:
∆z = tan λ0 · R0 · 2∆φ0 (14)
In the case of a positive turning circle R0 and ∆φ’ are both positive, in the case of a
negative turning circle R0 and ∆φ’ are both negative, so that the sign of ∆z depends only
on tan λ0 , as it should.
If D0 = 0 i.e. ∆φ0 = ∆φ and ρ0 = ρ one gets:
∆α
∆φ = (15)
2
For φ(φ/ρ) one gets:
ρ
sin(∆φ) = (16)
2R0
For φ(z) one gets:
∆z
∆φ = (17)
2R0 · tan λ0
From the equations 16 and 17 one gets for ρ(z):
∆z
ρ = 2R0 · sin (18)
2R0 · tan λ0

The equations 16 to 18 give the exact description of a helix in polar coordinates if D0 = 0.

If D0 6= 0 they are much more complicated: The equations for φ(ρ) are calculated in
appendix 1. With
 
D0
R1 = R0 + D0 D1 = D0 · 1 − (19)
2R1
one gets:
D1 ρ
sin(φ − φ0 ) = sin ∆φ = + (20)
ρ 2R1
The respective equations for φ(z) and ρ(z) are calculated in appendix 2 and 3:
D0 ∆z
tan(φ − φ0 ) = tan ∆φ = + tan (21)
∆z 2R0 · tan λ0
R0 · sin
R0 · tan λ0
 
2 D0 ∆z
ρ = D02+ 1+ · 4R02 · sin2 (22)
R0 2R0 · tan λ0
If D0 = 0, the equations 20 and 21 give back the equations 16 and 17.

5
4.2 Approximation in the inner and linear regions.
∆z
We assume D0  R0 and  R0 .
tan λ0
For φ(ρ) (equation 20) one gets::
D0 ρ
sin ∆φ ≈ + (23)
ρ 2R0
For φ(z) (equation 21) one gets::
D0 · tan λ0 ∆z
tan ∆φ ≈ + (24)
∆z 2R0 · tan λ0
For ρ(z) (equation 22) one gets::
 2
2 ∆z
ρ ≈ D02 + (25)
tan λ0

The border between the inner region and the linear region lies at the position, where the
2 terms of the addition in the 3 equations above are equal:
For φ(φ/ρ) (equation 23) one gets:
p
ρφ/ρ = 2R0 · D0 (26)

For φ(z) (equation 24) one gets:

p
∆zφ/z = tan λ0 · 2R0 · D0 (27)

For ρ(z) (equation 25) one gets for ρ(z):

∆zρ/z = D0 · tan λ0 (28)

Depending on the value of ρ or ∆z the left or the right term of the respective sum in the
equations 23 to 25 is dominant.
The border between the 2 regions is practically the same for φ(ρ) and φ(z), but much closer
to 0 for ρ(z).

4.3 Approximation in the inner region

For φ(ρ) one gets:
D0
sin ∆φ ≈ (29)
ρ

6
For φ(z) one gets:
D0 · tan λ0
tan ∆φ ≈ (30)
∆z

For ρ(z) one gets:

ρ ≈ |D0 | (31)

∆z
With ≈ ρ0 ( see equations 13 and 14 ) one can change the equations 25 and 30 to:
tan λ0
D0
tan ∆φ ≈ (32)
ρ0
ρ2 ≈ D02 + ρ02 (33)

The equations 29, 32 and 33 follow directly from figure 2c which replaces figure 2a in the
case R0 = ∞.

4.4 Approximation in the linear region

Most HEP detectors are built in a way, that the space points of tracks of interesting particles
lie on helix segments, which are only slightly curved up to the last measured space point.
The distance of the first measured space point to the beam line is significantly bigger than
the D0 of interesting tracks. Helix segments lie in the linear range as long as
D0 D0 · tan λ0
< Eφ , < Eφ , (34)
ρ ∆z
|∆φ − sin ∆φ| < Eφ and |∆φ − tan ∆φ| < Eφ (35)
where Eφ is the meassuring error of φ (see equations 23 and 24). For ρ(z) the range extends
∆z
to lower values of z, i.e. tan λ0
 D0 .
For φ(ρ) one gets then:
ρ
∆φ ≈ (36)
2R0

For φ(z) one gets:

∆z
∆φ ≈ (37)
2R0 · tan λ0

For ρ(z) one gets:

∆z
ρ≈ (38)
tan λ0

7
By replacing R0 and tan λ0 by PT and PZ (see equations 8 and 9) one gets:

ρ
∆φ ≈ s · e · B · (39)
2PT
∆z
∆φ ≈ s · e · B · (40)
2PZ
∆z PZ
≈ (41)
ρ PT

4.5 Approximation in the linear and outer region

The approximation in the linear and outer region are identical to the exact equations 16
to 18, which were derived for the case D0 = 0.

4.6 λ as a function of ρ
In the linear region of ρ(z) the polar angle λ0 is constant (see figure 3c and 3f).

∆z PZ
tan λ0 = = (42)
ρ PT
Several tracks may have different values of z0 , which may be due to measuring errors, to
multiple scattering or to a secondary vertex from which the track originates. In pattern
recognition the z-position zv of the primary vertex is determined, which lies on the z-axis,
i.e. xv = 0 and yv = 0. If the angle λ is seen from the primary vertex PV , it is no longer
constant close to the vertex (see figure 4).
z − zv
tan λ = (43)
ρ

From appendix 4 one gets:

z0 − zv
tan λ = tan λ0 + (44)
ρ

The behaviour is similar to the one of φ(ρ) close to the vertex, compare to equation 29.

8
5 Linearisation of φ(ρ) for visual pattern recognition
In HEP experiments it is sometimes of interest to identify points which belong to the same
helix, even in an environment of points belonging to other helices or of points due to noise.
We assume here that we are outside of the inner region, i.e. D0 and multiple scattering
are sufficiently small.
If the radius of such a helix is sufficiently large the points should lie on a straight line in
a φ/ρ, a φ/z or a ρ/z projection according to the equations 36, 37 and 38 and can be
identified as such.
If the radius is not large enough, they lie on a straight line in a φ/z projection (see equation
17). However, if one cannot approximate sin ∆φ by ∆φ, they ly on a more or less curved
line in the other projections (see equations 16 and 18). In the case of the φ/ρ projection
there is a method of linearisation:
Setting in Equation 16 sin(φ−φ0 ) = sin φ·cos φ0 −cos φ·sin φ0 and dividing by cos φ · cos φ0
one gets for the case D0 = 0:
1 ρ
tan φ ≈ · + tan φ0 (45)
2R0 · cos φ0 cos φ
If one rotates the coordinate system by an arbitrary angle -φ1 one gets:
1 ρ
tan(φ − φ1 ) ≈ · + tan(φ0 − φ1 ) (46)
2R0 · cos(φ0 − φ1 ) (cos φ − φ1 )
ρ
If one makes a projection of tan(φ − φ1 ) versus one gets again a straight line.
cos(φ − φ1 )
If one sets φ1 ≈ φ0 or to the φ value of one of the points in question tan(φ − φ1 ) is not
ρ
very different from φ − φ1 and is not very different from ρ. Therefore, this
cos(φ − φ1 )
projection looks similar to the φ/ρ projection, except that the points of the helix ly now
on a straight line.

6 Determination of the perigee paramters

From the cylindrical coordinates of 2 points on a circle one can calculate the perigee pa-
rameters from equations 36 or 37 and 38, if the points lie in the linear region. In that case
D0 cannot be calculated and is set to 0. R0 , φ0 and z0 are found by solving a set of linear
equations.
If the points lie in the linear and the outer region R0 and φ0 can be calculated by solving
linear equations, via equation 45 and D0 is set to 0.
If one wants to calculate D0 as well, one needs 3 points of which at least one should be
sufficiently close to the inner region, so that its offset from a straight line is larger than
the measuring errors.

9
In order to calculate R0 , φ0 and D0 by use of linear equations, one needs to modify equation
20 in the same way equation 45 was deduced.
Setting in Equation 20 sin(φ−φ0 ) = sin φ·cos φ0 −cos φ·sin φ0 and dividing by cos φ · cos φ0
one gets:
D1 1 1 ρ
tan φ = tan φ0 + · + · (47)
cos φ0 ρ · cos φ 2R1 · cos φ0 cos φ

D1 1
Solving this equation for 3 points one gets tan φ0 , and from which one
cos φ0 2R1 · cos φ0
can calculate R0 and φ0 and D0 .
To calculate tan λ0 and z0 one can use equation 38, with a linear region reaching far down
to 0. Two points are sufficient, which, however, must not lie in the outer region.

7 Conclusions
The calculation of space points along a circle or helix close to the point of nearest approach
may lead to numerical problems if the radius R0 is very large and one uses the simple
carthesian equations 6, because one calculates small numbers by subtracting very large
ones. This problem is avoided if one calculates the cylindrical space points via equations
19 to 22.
As already stated in section 4.4, most HEP detectors are built in a way, that the space
points of tracks of interesting particles lie on helix segments, which are only slightly curved
up to the last measured space point. As the distance of the first measured space point
to the beam line is normally significantly bigger than the D0 of interesting tracks, the
measured space points lie in the linear range of φ(ρ), φ(z) and ρ(z). These functions are
described by the very simple set of linear equations 36 to 41.
The linear behaviour of these function leads to helix projections φ/ρ, φ/z, ρ/z and φ/λ,
which ease considerably detection and analysis of particle tracks through their measured
space points.
The deviations from linearity are due to the approximation of sin ∆φ by ∆φ and a behaviour
1 1
of φ(ρ) and φ(z) very close to the vertex described by and respectivley.
ρ ∆z

8 Appendix 1: φ(ρ)
From figure 2a one gets:

U = ρ · cos ∆φ V = ρ · sin ∆φ (48)

10
 U = R0 · sin ∆α V = R0 + D0 − ρ · cos ∆α (49)
ρ · cos ∆φ = R0 · sin ∆α R0 + D0 − ρ · sin ∆φ = R0 · cos ∆α (50)
with

R1 = R0 + D0 (51)

one gets by squaring and adding the equations 50:

R12 − 2R1 · ρ · sin ∆φ + ρ2 · sin2 ∆φ + ρ2 · cos2 ∆φ = R02 (52)

R12 − R02 + ρ2 = 2R1 · ρ · sin ∆φ (53)

R12− R02 ρ 2
+ = sin ∆φ (54)
2R1 · ρ 2R1 · ρ
With
R12 − R02 R2 − R12 + 2R1 · D0 − D02
D1 = = 1 (55)
2R1 2R1
 
D0
D1 = D0 · 1 − (56)
2R1
one gets the final result:
D1 ρ
+ = sin ∆φ (57)
ρ 2R1

9 Appendix 2: φ(z)
Equations 14 and 13 are used below:

∆z ρ0
= 2R0 · ∆φ0 = sin ∆φ0 (58)
tan λ0 2R0
From figure 2a one gets:

U = ρ0 · cos ∆φ0 V = D0 + ρ0 · sin ∆φ0 (59)

V D0 + ρ0 · sin ∆φ0
tan ∆φ = = (60)
U ρ0 · cos ∆φ0
Replacing ρ’ via equation 58 (right) one gets:

D0
tan ∆φ = 0 0
+ tan ∆φ0 (61)
2R0 · sin ∆φ · cos ∆φ

11
 D0
tan ∆φ = + tan ∆φ0 (62)
R0 · sin 2∆φ0
This equation gives ∆φ as function of ∆φ’. Replacing ∆φ’ via equation 58(left) one gets
the final result:
D0 ∆z
tan ∆φ = + tan (63)
∆z 2R0 · tan λ0
R0 · sin
R0 · tan λ0

10 Appendix 3: ρ(z)
Equations 14 and 13 are used below (as equations 58):
∆z ρ0
= 2R0 · ∆φ0 = sin ∆φ0 (64)
tan λ0 2R0
From figure 2a one gets(as equations 59):

U = ρ0 · cos ∆φ0 V = D0 + ρ0 · sin ∆φ0 (65)

ρ2 = U 2 + V 2 = ρ02 + D02 + 2D0 · ρ0 · sin ∆φ0 (66)

Replacing sin ∆φ’ via equation 64 (right) one gets:

2D0 · ρ02
ρ2 = D02 + ρ02 + (67)
2R0
 
2 2 02 D0
ρ = D0 + ρ · 1 + (68)
R0
Replacing ρ’ via the equations 64 one gets the final result:
 
2 2 D0 ∆z
ρ = D0 + 1 + · 4R02 · sin2 · (69)
R0 2R0 · tan λ0

11 Appendix 4: λ(ρ)
From figure 4 one gets:
z − z0 z − zv
= tan λ0 = tan λ (70)
ρ ρ
From this follows the final result:
z0 − zv
tan λ = tan λ0 + (71)
ρ

12
12 Appendix 5: Programming example
subroutine test(r0,d0,fi0,z0,alambda0,dalpha)
c --------------- <<<<<<<<<<<<<<<<<<<<<<<<<<<<
c ........................... Input parameters
c Test equations:

character *20 terr

call store_helix_parameters(r0,d0,fi0,z0,alambda0)

call get_circle_center(xc,yc)
call get_circle_pd(xd,yd)

call get_point_at_dalpha(dalpha,x1,y1,z1,terr)
if(terr.ne.’ ’) go to 9

if(dalpha.eq.0.) then
rho=d0
rhoprime=0.
else
rho=sqrt(x1*x1+y1*y1)
rhoprime=sqrt((x1-xd)*(x1-xd)+(y1-yd)*(y1-yd))
end if

call get_point_at_rho(rho,x2,y2,z2,terr)
if(terr.ne.’ ’) go to 9

call get_point_at_rhoprime(rhoprime,x3,y3,z3,terr)
if(terr.ne.’ ’) go to 9

c ......................................................... Output
write(*,1000) ’xc,yc,xd,yd’,xc,yc,xd,yd
1000 format(1x,a,4f14.2)
c ............................ The 3 sets of x,y,z should be equal.
write(*,1001) ’x1,y1,z1’,x1,y1,z1
write(*,1001) ’x2,y2,z2’,x2,y2,z2
write(*,1001) ’x3,y3,z3’,x3,y3,z3
1001 format(1x,a,3e20.8)
return

9 write(*,9000) terr

13
9000 format(1x,a)

stop
end

subroutine store_helix_parameters(r0_i,d0_i,fi0_i,z0_i,alambda0_i)
c -------------------------------- <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
c .......................................... Store input parameters.
c ............................................ all angles in radian.
c .............................. r0>0 and alambda0_i # pi/2 assumed.
character *20 terr
r0=r0_i
d0=d0_i
fi0=fi0_i
z0=z0_i
alambda0=alambda0_i
c ................................................ equations 4
xc=-(r0+d0)*sin(fi0)
yc= (r0+d0)*cos(fi0)
c ................................................ equations 5
xd=-d0*sin(fi0)
yd= d0*cos(fi0)
c ................................................ equations 19
r1=r0+d0
d1=d0*(1.-(d0/(2.*r1)))
c ................................................ used below
rtl=r0*tan(alambda0)
return

entry get_circle_center(xc_o,yc_o)
c ----------------------- >>>>>>>>>
c ................................................ equations 4
xc_o=-xc
yc_o= yc
return

entry get_circle_pd(xd_o,yd_o)
c ------------------- >>>>>>>>>
xd_o=xd
yd_o=yd
return

entry get_point_at_dalpha(dalpha,x,y,z,terr)

14
c ------------------------- <<<<<< >>>>>>>>>>
c ........... all angles in radians. if r0<0 then dalpha<0 !
if(r0*dalpha.lt.0.) then
terr=’r0*dalpha.lt.0.’
else
terr=’ ’
c .............................................. equations 6,7
x=xc+r0*sin(fi0+dalpha)
y=yc-r0*cos(fi0+dalpha)
z=z0+tan(alambda0)*r0*dalpha
terr=’ ’
end if
return

entry get_point_at_rho(rho,x,y,z,terr)
c ---------------------- <<< >>>>>>>>>>
if(rho.lt.d0) then
terr=’rho.lt.d0’
else
terr=’ ’
c ............................................... equation 20
if(rho.eq.d0) then
sindf=1.
else
c ...... If(rho=d0), sindf may be numeriacally just bigger 1.
sindf=d1/rho+rho/(2.*r1)
end if
df=asin(sindf)
x=rho*cos(fi0+df)
y=rho*sin(fi0+df)
c ............................................... equation 22 inversed
sq=sqrt((rho*rho-d0*d0)/(1.+d0/r0))
as=asin(sq/(2.*r0))
dz=2.*rtl*as
z=z0+dz

end if
return

entry get_point_at_rhoprime(rhop,x,y,z,terr)
c --------------------------- <<<< >>>>>>>>>>
if(rhop.lt.0) then
terr=’rhop.le.0’

15
 else
terr=’ ’
c ............................................... equation 16
sindf=rhop/(2.*r0)
df=asin(sindf)
xp=rhop*cos(fi0+df)
yp=rhop*sin(fi0+df)
x=xd+xp
y=yd+yp
c ............................................... equation 18 inversed
as=asin(rho/(2.*r0))
dz=2.*rtl*as
z=z0+dz

end if
end

16
 H1 Circle equations in carthesian coordinates
H.Drevermann: Helix equations

Pc
Y − Yc

∆α

φo P
− Y

Ro

o
90

φo
Pd − Yd

Filename: ed_h1.ps
Made on 2-Apr-2007 09:39:21 by hansd with ATLANTIS
Do
φo
o
90
X
| | Po |
Xc Xd X
1)

Figure 1: ”Circle equations in carthesian coordinates”

The pdf file ed bw h1.pdf contains the figure in black and white.
The pdf file ed co h1.pdf contains the figure in color. When displaying the color picture
the quality is better, if acroread is used.

17
 H2 Circle equations in cylindrical coordinates H.Drevermann: Helix equations

90 deg
Ro
∆α ∆φ’

∆φ’
P
U

ρ’

Ro
ρ

Filename: ed_h2.ps
Made on 2-Apr-2007 09:39:28 by hansd with ATLANTIS
φ

∆φ’
0

Do 0 ρ 800mm
2b)
Y
Do ρ’ ∆φ
φ
∆φ ρ

φ 1/R o = 0
o ∆φ
2a) X 2c)

Figure 2: ”Circle equations in carthesian coordinates”

The pdf file ed bw h2.pdf contains the figure in black and white.
The pdf file ed co h2.pdf contains the figure in color. When displaying the color picture
the quality is better, if acroread is used.

18
 H3 φ / ρ , φ / z , ρ / z , sin φ / ρ H.Drevermann: Helix equations

R o = 1.67 m, Do = 0.5 mm R o = 1.67 m, Do = 0 mm R o = 167 m, Do = 0.5 mm

2000mm
1

1
ρ

φ φ
0

0
∆Z

Filename: ed_h3.ps
Made on 2-Apr-2007 09:39:33 by hansd with ATLANTIS
3a)
0 ρ 2000mm
3b)
0 ∆Z 2000mm
3c)
0 2000mm

4mm
1

ρ
φ φ
0
0

3d)
0 ρ 40mm
3e)
0 ∆Z 40mm
3f)
0 ∆Z 4mm

Figure 3: ”φ/ρ, φ/z and ρ/z projection of aa helix”

The pdf file ed bw h3.pdf contains the figure in black and white.
The pdf file ed co h3.pdf contains the figure in color. When displaying the color picture
the quality is better, if acroread is used.

19
 H4 λ as seen from the vertex H.Drevermann: Helix equations

ρ P

λo
λ

Filename: ed_h4.ps
Made on 2-Apr-2007 09:39:40 by hansd with ATLANTIS
λo λ

Po Pv Z
0 Zo Zv Z

4)

Figure 4: ”λ as seen from the vertex”

The pdf file ed bw h4.pdf contains the figure in black and white.
The pdf file ed co h4.pdf contains the figure in color. When displaying the color picture
the quality is better, if acroread is used.

20