
3
renormalize the charge’s inertia. Even for smooth charge
models without divergences, coupling of the Maxwell and
Lorentz equations alone implies a change of the inertial
bare mass of the charge [3,4], which needs to be gauged
to the experimentally measured one.
For the DSR equation we proposed the following renor-
malization scheme in [1]. As the geometry of the tube,
recall Figure 1, is not static but dynamic, one may expect
some of the contributions to the inertia mto depend on
the parameter τ. Therefore, we start with the ansatz
pα(τ) = m(τ)uα(τ).
Exploiting uα(τ)aα(τ) = 0 for for world lines, we identify
d
dτ m(τ) = Lα
(τ)uα(τ).(4)
Setting
Fα
(τ) = Lα
(τ)−Lβ
(τ)uβ(τ)uα(τ),
Fα
ext(τ) = qF αβ(z(τ))uβ(τ)
results in relativistic four-forces, i.e., ones being four-
perpendicular to the four-velocity of the charge. Finally,
we find the following system of equations
d
dτ
zα(τ)
uα(τ)
m(τ)
=
uα(τ)
1
m(τ)(Fα
(τ) + Fα
ext(τ))
uα(τ)Lα
(τ)
,(5)
which we refer to as the DSR system of equations.
For > 0, the DSR system (5) represents a neutral
system of delay differential equations, meaning that the
highest derivative of a solution component occurs with
and without delay. A general solution theory, and espe-
cially a study of the case →0, is still pending. Nev-
ertheless, candidates for initial conditions are certainly
twice continuously differentiable segments of the charge
trajectory
[−, 0] →R4, τ 7→ zα(τ)
together with an initial inertia m(0). The static divergent
inertia in (3) can be renormalized by setting the initial
inertial mass equal to
m(0) = m−q2
2,(6)
which at the same time sets the effective initial inertia to
some experimental value m(0). Note however, that m(τ)
may still be both τ- as well as -dependent. Nevertheless,
there is numerical evidence that the corresponding solu-
tions of the DSR system depend only weakly on which
also raises the hope that a rigorous analysis of the limit
→0 can be done. While we think, this is the case, it
must be noted that the right-hand side of (6) becomes
negative for below the classical electron radius and, in
this case, renders the corresponding dynamics unstable;
as already seen in [4] for the coupled Maxwell-Lorentz
system of rigid charges.
B. A reduced DSR model for small
In this section, the goal is to formulate a simplified, yet
still meaningful, version of the system (5), namely with
the force (8) below, from which we can expect that the
corresponding solutions sets are close in certain regimes
and -ranges. The main advantage of (8) is the much
simpler structure, which reduces the numerical cost sig-
nificantly.
This simplified force can be motivated by regarding
only the following, in the limit →0 for fixed τleading,
terms:
Lα
(τ)≈ − q2
2aα+2q2
3(aα(τ)−aα(τ−))
+2q2
3aγaγuα.(7)
Now we permit us to drop all high-order terms and even
remove the part of the force in uα-direction in order to
avoid a τ-dependend inertia coming from (4) which re-
sults in a constant m(τ) = m−q2
2. The lower two com-
ponents of (5) can now be combined into the equation
maα=Fα
ext +2q2
3aγ(τ−)uγuα
+2q2
3
aα(τ)−aα(τ−)
.
(8)
We will refer to the system of equations comprised by
the first row of (5) and equation (8) together with the
above constant choice of m(τ) as the reduced DSR sys-
tem. Note the additional advantage that the right-hand
side of (8) is linear in the acceleration so that the corre-
sponding flow can be given explicitly in contrast to the
one of (7). In (8), we will again assume > 2q2
3m(19) to
avoid the discussed dynamical instability.
In view of our objectives in formulating the reduced
DSR system, the numerical cost evaluating (8) is now
considerably lower as compared to the full DSR model
(5), i.e., (17) below, or even the LL model, c.f., (A1)
below, as for the latter, the derivative of the field strength
tensor, (A12) below, is needed in the multi-particle case.
A similar equation has been suggested in [5]. Equations
(8) still lead to a change of the inertia mas can be seen
in the case of uniform acceleration. Say, Fα
ext was tuned
such that it produces a solution with four-velocity taking
the form uα= (cosh(gτ),sinh(gτ),0,0), we find
2q2
3aγ(τ−)uγuα−2q2
3
aα(τ−)−aα(τ)
=−4q2
3sinhg
22aα
similar to the change in inertia in the case of the DSR
system discussed in [1]. This illustrates also that m(τ)
may not simply be regarded as the total inertia of the
charge.