LOSSES IN FERRITE ROD ANTENNAS

Abstract

The losses in ferrite rod antennas are much higher than predicted by the accepted theory. This article shows that the increased loss is due to an increase in the copper losses in the winding rather than losses in the ferrite itself.

Full text

Payne : Losses in Ferrite Rod Antennas, Issue 2
1
LOSSES IN FERRITE ROD ANTENNAS
The losses in ferrite rod antennas are much higher than predicted by the accepted theory.
This article shows that the increased loss is due to an increase in the copper losses in the
winding rather than losses in the ferrite itself.
1. INTRODUCTION
When a ferrite rod is introduced into a coil its inductance increases considerably, and if the only increase
in loss was in the ferrite a very large increase in Q would result. But experiment does not support this, and
one of the first to notice this was Polydoroff (1960, ref 1) who found that ‘the increase of loss due to the
insertion of the core is many times greater than the iron loss, proper’ (he was using dust iron cores). In one
of his experiments the loss resistance introduced by the magnetic core was nearly 100 times the value he
had calculated, but despite an extensive investigation he was unable to locate the additional loss.
Surprisingly the situation has not improved much over the subsequent half century, and textbooks still
assume that the all the additional loss is in the magnetic core. For instance the ‘Antenna Engineering
Handbook’, Johnson & Jasik (ref 2, Chapter 5) gives the ‘resistance due to the core loss’ as :
Rm= ω (μrod / μ’) 2 μ’’μ0 Fr N2 A/ lc 1.1
Where μ’ is the ferrite permeability, μ’’ its loss permeability, N is the number of turns, A is the cross-
sectional area of the ferrite and lc the coil length. For the calculation of μrod the theory of demagnetisation is
used, derived from the theory of permanent magnets (see Payne ref 4). Fr is an empirical factor needed to
get the equation to agree with experiment, and this ranges from about 0.1 to 0.6 but even then the equation
is not accurate because Fr is derived from ’averages of experimental data’ (ref 2).
The reason for this inaccuracy is in the assumption that the wire resistance remains the same as in the air
cored coil, and that the only additional loss when a ferrite core is introduced is that in the ferrite itself.
However the wire loss increases considerably, and this is the topic of this article.
The total loss is the sum of the conductor loss and the ferrite loss, and these are considered below starting
with the ferrite loss.
In this report the most significant equations are highlighted in colour.
2. FERRITE LOSS
2.1. Closed core and Open core
Most magnetic circuits are termed ‘closed’, in that the ferrite forms a complete magnetic circuit containing
the entirety of the flux, and the most common example is the transformer. Sometimes an air gap is included
but this is generally very short so that any leakage flux is very small. In contrast the ferrite antenna is an
‘open’ magnetic circuit having a very large leakage flux, but the same theory can be used to evaluate the
ferrite losses, so the analysis below starts with closed cores.
2.2. Losses in Closed Cores
The permeability of a ferrite is designated by μ’ and its loss and by an imaginary permeability μ’’. These
both change with frequency and so are often given by the manufacturer as a graph similar to the one below:

Payne : Losses in Ferrite Rod Antennas, Issue 2
2
Complex Permeability vs Frequency
0.1
1
10
100
1000
0.1 1 10 100
Frequency Mhz
Permeability
μ'
μ''
The Q of a ferrite is given by the ratio of these two:
Qm = μ’ / μ’’ 2.2.1
In the graph above the Q of the ferrite at10 MHz is Qm = 57 (i.e. 170/3).
If a coil is wound around a closed ferrite, such as a toroid, and measurements made of the inductance and
resistance then the overall Q is equal to the ratio of reactance to resistance Q = ω L/ (Rw + Rf), where Rw is
the loss due to the wire and Rf that due to the ferrite. If the wire resistance is subtracted this gives the Q of
the magnetic circuit (the ferrite) which is designated here as Qm :
Qm = ω L/ Rf 2.2.2
[NB for a closed magnetic circuit such as a toroid, it is assumed that the wire loss is not affected by the
inclusion of the ferrite. This seems to be essentially true when the permeability of the ferrite is high, so that
any leakage flux from the ferrite cutting the wire is very small. However, there is evidence that if the
permeability is small this may not be true, and then the wire loss can increase (see Polydoroff ref 1, p72)].
Equating the above two equations gives :
Qm = μ’ / μ’’ = ω L/ Rf 2.2.3
So ferrite manufacturers usually determine μ’ and μ’’ by winding a coil around a toroid of the material and
measuring the increase in inductance and the increase in resistance due to the core.
2.3. Losses in Closed Cores with an Air Gap
When an air gap is included in the magnetic circuit, the overall permeability reduces and so does the loss. If
the new permeability is μ’e , then the new loss factor is (Snelling ref 3, equation 4.48) :
Qm gapped = Qm. (μ’-1)/ (μ’e -1) 2.3.1
As an example, if a ferrite toroid with a permeability μ’ =200, and Qm =100, has inserted into it an air gap
of such a width as to reduce the effective permeability μ’e to say 50, then the new Q of the magnetic circuit
will rise to Qmgapped= 406.

Payne : Losses in Ferrite Rod Antennas, Issue 2
3
At first sight this result may be surprising, since by adding an air gap it might be expected that both μ’ and
μ’’ will reduce by the same amount, so that the Q would be unchanged. This would be true if the air gap
was also lossy with the same loss as the ferrite, but the air gap has no loss.
2.4. Losses in Open Cores
Considering now the ferrite rod antenna. When the rod is inserted into the coil the inductance increases by
an amount significantly less than the permeability μ’, because of the diluting effect of the large air gap. For
instance the inductance may only increase by a factor of 6 times even though the ferrite has a permeability
of say 250. This ratio by which the inductance actually increases, Lf/Lair , is often given the label μcoil, so to
apply Equation 2.3.1 to the open core we substitute μcoil for μe :
Qm open = Qm (μ’-1)/ (μcoil -1) 2.4.1
The ratio (μ’-1)/ (μcoil -1) is normally very large and in the absence of any other loss a large increase in Q
could be expected. So if the ferrite material in the above example had a Q of 100 then the Q of the ferrite
rod antenna would be 5000 [ie = 100*(250-1)/(6-1)].
2.5. µcoil
The above equation needs a value for μcoil, but the accepted equations for this are inaccurate. The author has
derived an accurate equation (Payne ref 4) which can be approximated to :
Lf/Lair ≈ x (1+C’) 2.4.2
where x = 5.1 [l ’ / dc ]/[1+ 2.8 (dc/ l ’ )]
l ’ = lc + 0.45 dc
C’ ≈ 0.7 [(l f - l c ) / dc ] / [ Ln {2 (l f + df)/ df } -1]
This equation should be accurate enough for practical purposes if the antenna meets the following
conditions:
a) the coil is no longer than twice its diameter and is centred on the ferrite rod.
a) the inner winding radius of the coil is not too different from that of the ferrite
b) the ferrite permeability is greater than 100
c) the ferrite rod is no longer than 12 times its diameter.
In applying this equation it should be noted that in the author’s experience the permeability of ferrite rods is
around half that quoted by manufacturers. For the measurement of the permeability of ferrite rods see ref 4.
2.6. Ferrite loss alone
Equations 2.4.1 and 2.4.2 give the loss in the ferrite, but this is only a fraction of the total loss as the
following experiment shows : a coil of 32 turns of 0.6mm dia enamelled copper wire, was wound over a
length of 20mm on a former of 10 mm dia., and gave a Qc of 156 at 3Mhz. When a ferrite rod was
introduced having a length equal to that of the coil, the above equations give a Q of 615, but the measured
Q was only 62.
[Details of experiment were : When the ferrite was introduced the inductance increased by 6.38 (= μcoil =
Lf/Lair). So if there was no loss in the ferrite and the winding loss was unaffected, the Q would increase to
995 (ie 156 x 6.38). In fact the Q of the ferrite was 29, and its permeability 300, both at 3 MHz. So from
Equation 2.4.1 the Q of the open core Qm open was 1612.
We could therefore expect the overall Q to be that of a Q of 995 in parallel with a Q of 1612, so
Qt = Q1 Q2/(Q1+Q2) = 615, but measurements gave a Q of only 62].
So the loss in the ferrite was only a small fraction of the total loss and we need to look for a large additional
loss.

Payne : Losses in Ferrite Rod Antennas, Issue 2
4
3. POSSIBLE LOSS MECHANISMS
In an effort to locate the extra loss the author considered many possible loss mechanisms, but most were
discounted because if present they would also be present in a toroid, and since this is used to determine the
loss in the ferrite Qm = μ’ / μ’’, any such loss would already be included. The loss mechanism had to be
unique to the open core, and not present in the toroid.
Two mechanisms were good candidates. The first was that there was a circulating flux around the core.
This could arise because it has been reported that at high frequencies in open cores the flux tends to form a
tube, with little flux in the centre of the core (Polydoroff ref 1 page 2).When flux traveling down this outer
tube arrives at the end of the ferrite rod it will not necessarily flow into the air, because there is a much
lower reluctance path it can take. This path is across the end face of the ferrite and then in the reverse
direction down the center core of the ferrite. This could lead to a large re-circulating flux giving high
losses, with a much smaller flux propagating into the air. This seemed an excellent candidate and so was
investigated in great depth, but its predictions did not correspond with measurements.
The second candidate was increased loss in the wire due to the large increase in the fields cutting the wire,
and the theory developed for this mechanism did agree with experiment and it is this which is described in
this article.
4. INCREASED COPPER LOSSES
4.1. Copper Losses at High Frequencies
So consideration must be given to the losses in the wire and how they are affected by the presence of the
ferrite.
It is well known that the resistance of isolated conductors at high frequency is much larger than the DC
resistance due to the skin effect, so-called because at sufficiently high frequencies the current penetrates
into the conductor only a small amount and flows in only a thin skin down the outside of the conductor.
Skin effect arises because of the magnetic field which the conductor produces around itself, but if this field
also intercepts other conductors this will also increase its resistance. In particular if the wire is wound into a
coil, each turn of the wire induces loss-making eddy currents into adjacent turns, indeed even of turns some
distance away in that it contributes to an overall field down the coil. The power lost in these eddy currents
must come from the wire(s) responsible for the magnetic field, and so these wires have an apparent increase
in their resistance. In addition, as flux passes down the coil only a part of it reaches the end, and a
proportion leaks away into the wire, particularly towards the ends of the coil, and this sets up further eddy
currents. In Figure 4.1 this leakage flux is shown exiting the coil, but as it cuts the wires the resultant eddy
currents tend to cancel this flux. This cancellation is especially pronounced at high frequencies and with
close wound coils, and then little if any flux exits from the sides.

Payne : Losses in Ferrite Rod Antennas, Issue 2
5
Figure 4.1 Flux Distribution in an Air Coil
4.2. Loss due to Magnetic Field
The apparent resistance of the wire therefore increases, and if this extra resistance is called Rφ, then the
power lost in the coil is:
PT= I2 R0 + I2 Rφ 4.1.1
The first term is the power lost in the wire when the external field is absent. The second power loss is due
to the magnetic field cutting the wire, and if this has an intensity H amps/m the power loss due to this cause
is (Welsby ref 5 p53):
Wφ = I2 Rφ = 4π2 RDC d2 H2 G watts 4.1.2
Where RDC is the dc resistance of the wire, G a numerical factor which allows for the skin depth, and d the
diameter of the wire.
The additional power loss is therefore proportional to H2, and when the ferrite is introduced H increases by
Lf/Lair and so the power loss increases as (Lf/Lair)2. Since this extra power loss occurs without an increase in
the current I, it must come from an apparent increase in the resistance, and so the resistance of the wire will
increase by (Lf/Lair)2.
Therefore to a first approximation the wire loss increases by (Lf/Lair)2, and this goes a long way to
explaining Polydoroff’s additional loss described in the Introduction.
5. CONDUCTOR LOSS IN AIR COILS
The inclusion of the ferrite therefore increases the resistance of the air coil, and so it is air coils which are
considered first.
The alternating current resistance of single layer air coils wound with round wire presents serious
mathematical difficulties and was attempted by Butterworth in the 1920’s, but his analysis has been shown
to be not very accurate for the important case of close wound coils (Medhurst ref 6). In 1951 a more

Payne : Losses in Ferrite Rod Antennas, Issue 2
6
accurate theory was developed by Arnold (ref 7), but again the analysis is complicated and the resulting
equations difficult to use.
More recently the author has produced a new analysis (ref 8) which leads to the following equation for the
resistance of the wire, and this equation has the advantage that it can be readily programmed into spread-
sheets:
Rwire = Row + Row kr Kn2 + Rwall N2 32 π (1- Kn ) (dw /p)av M2 (le / lc)2 (acoil / l coil )/ (w2 / w1) 5.1
{The individual factors are defined later}.
The above equation is of the form :
Rwire = Row + RAw + RRw 5.2
Row is the high frequency resistance of the unwound wire conductor, and RAw and RRw are the added
resistances due to the axial and radial components of the fields cutting the conductor (see Figure 5.1).
Figure 5.1 Axial and Radial Fields
When the ferrite is added these axial and radial fields are changed, and this changes the values of RAw and
RRw as detailed below.
6. EQUATIONS MODIFIED FOR FERRITE ANTENNAS
6.1. Flux Distribution in Ferrite Cored Coil
When the ferrite is introduced into the air coil the main effect is that the magnitude of the flux increases by
Lf/Lair, and if this were the only change then the loss components RAw and RRw would be increased by
(Lf/Lair)2 [see Section 4]. However, the shape of the field also changes, as shown in Figure 6.1.

Payne : Losses in Ferrite Rod Antennas, Issue 2
7
Figure 7.1 Flux Distribution in Ferrite Coil
The key difference in the shape is that the magnetic flux takes the course of least reluctance which is
through the ferrite, so the axial field through the wires is very much smaller than in the air coil. The effect
of this change is considered below.
6.2. The Conductor Loss due to the Axial Field
In the air coil the current in the wire is concentrated onto the side of the wire closest to the centre of the
coil. This arises because the axial field is stronger on the inside of the coil than on the outside (Figure 4.1),
and so a filament of the wire on the outside is enclosed by more flux than a filament of wire on the inside.
The outside filament therefore has a higher inductance and its current is reduced.
The axial field through the wire is reduced considerably when the ferrite is introduced, and indeed if the
ferrite permeability is high it can be assumed that the field is close to zero, and that current flows over the
whole circumference of the wire. Thus RAwf ≈ 0 for high permeability. Notice that this is a reduction in
resistance.
There are two other factors in Equation 5.1 which are dependent on the axial field : (dw /p)av and (w2 / w1).
The first of these is apparent ratio of the wire diameter to the pitch, and this differs from the actual ratio
because of currents induced in the wire by the axial field. When the axial field is zero (dw /p)av = (dw /p).
The second factor (w2 / w1) becomes unity when there is no axial field.
6.3. The Conductor Loss due to the Radial Field
When the ferrite is introduced the loss resistance RRw can be expected to increase to :
RRwf ≈ RRw (Lf/Lair)2 6.3.1
The coil diameter will be slightly greater than that of the ferrite, and so the flux leaving the ferrite will have
a lower density when it reaches the coil, by the ratio of their areas (ac/af)2. So Equation 7.3.1 becomes :
RRwf ≈ RRw (Lf/Lair)2 (ac/af)2 6.3.2
There is also another adjustment to be made : those factors in Equation 5.1 which describe the intensity and
shape of the flux [Kn, M and le / lc ] are now describing this within the ferrite (over the length of the coil)

Payne : Losses in Ferrite Rod Antennas, Issue 2
8
and so in the calculation of these it is the ferrite diameter which must be used rather than that of the coil. In
this respect it is as if the coil was projected onto the surface of the ferrite.
6.4. Loss in the Ferrite
The series loss resistance due to the ferrite itself has to be added to the above conductor loss. The Q of the
open ferrite is given by Equation 2.4.1, so the series resistance due to this is loss is given by
Rf = ω Lf/ (Qm open) 6.4.1
where ω =2πf
Lf is the inductance with ferrite
Qm open is given by Equation 2.4.1
7. Overall Loss Equation with Ferrite Core
From the previous section the overall resistive loss is given by :
Rwire t = Row + RRw (Lf/Lair)2 (ac/af)2 + ω Lf/ (Qm open) 7.1
where Lf/Lair is given by Equation 2.4.2
Lf is the inductance with ferrite (Payne ref 4)
l coil is the length of the winding (see para 8.1)
Qm open is given by Equation 2.4.1
Row is the loss of the unwound straight wire, and if (2π acoil N)2 >> l coil 2 (the normal case ) is given by (see
Payne ref 8) :
Row ≈ Rwall 2 acoil N2 / [l coil (dw /p)] 7.2
where dw is the effective wire diameter (see below)
p is the pitch of the winding
N is the number of turns
acoil is the radius of the winding to the centre of the wire
Rwall = ρ / δ where ρ is the resistivity and δ the skin depth of the
conductor. For copper Rwall = 0.264 10-3 f 0.5, where f is in MHz
Raw is the loss due to the axial field and is given by :
RRw = Rwall N2 32 π (1- Kn ) (dw /p) M2 (le / lc)2 (acoil / l coil ) 7.3
where Kn ≈ 1/ [ 1+ 0.45 (df / lcoil) - 0.005 (df / lcoil)2 ]
M ≈ df /[ (2 df)2 + ( lcoil) 2] 0.5
(le / lc) ≈ Kn (1+ 0.05 df / lcoil )
df is the diameter of the ferrite
The effective diameter of the wire is smaller than its physical diameter because the current recedes from the
surface by one half of the skin depth (see Wheeler ref 9). This effect can be very large, and for instance a
wire with a physical diameter of 0.2 mm will have an effective diameter 33% less at 1 MHz.

Payne : Losses in Ferrite Rod Antennas, Issue 2
9
So dw = d- δ, where d is the physical diameter and δ the skin depth which for copper is equal to 66.6 /f 0.5
where f is in Hz. This is accurate to within 5.5% for wire diameters greater than twice the skin depth (dw / δ
> 2).
8. EXPERIMENTAL RESULTS
8.1. General
In the following a comparison is made between experiment and the theory given above. To avoid any errors
in the calculation of the ratio Lf/Lair, this ratio was derived from the measurements, and used in the
calculation of loss resistance.
The calculated resistance is very sensitive to the values of the coil diameter and length, and the diameter of
the wire, and this sensitivity is discussed later. The definition used here for these parameters is : the wire
diameter is the physical diameter minus the skin depth. Skin depth is 0.036 mm in copper at 3.4 MHz and
the wire diameter was measured at 0.55 mm after burning off the insulation.The coil diameter is the mean
diameter of the winding ie the diameter of the former on which the wire was wound plus one wire diameter
The coil length is N times the pitch, where N is the number of turns. Notice that this is greater than the
distance from the first turn to the last turn and represents the equivalent length of the current sheet (see
Grover ref 10, p149).
8.2. Comparison of Theory and Experiment
A coil of 17 turns was wound onto a low loss former with 0.55mm dia wire, over a current sheet length of
22.6mm, so the ratio dw/p was 0.39. The mean winding diameter to the centre of the wire was 11.84mm. Its
air inductance was measured at 1.55 µH, and corrected at each frequency for its self-resonant frequency
SRF, using Equation 9.1.1. (SRF was measured as 75 MHz). This corrected value of Lair was used to
calculate the ratio Lf/Lair.
Nickel- Zinc ferrite rods of various lengths up to 120mm were inserted into this coil, each rod having a
diameter of 10mm, but with two flats along the sides. The effective diameter was estimated at 9.42 mm.
At each length the inductance, series resistance and SRF of the coil were measured at 3.4 MHz, and the
values corrected for the SRF (measured for each rod length) according to Equations 9.1.1 and 9.1.2
The frequency of 3.4 MHz was chosen because the Q of this particular ferrite was only 35 at this frequency
(see below), and thus the ferrite would give a fairly high loss resistance, enabling this aspect to be
evaluated.
In the graph below the measurements of the total loss resistance are shown in blue and the calculated total
loss (Equations 7.1, 7.2 and 7.3) is in red. Also shown in green is the calculated contribution to the total
loss by the ferrite (Equation 6.4.1):

Payne : Losses in Ferrite Rod Antennas, Issue 2
10
Figure 8.2.1 Comparison of Measured and Calculted Resistance
The agreement is very good given the possible measurement errors, and the need to extrapolate the ferrite
parameters to this frequency (see below). Notice that the ferrite loss is a small part of the total loss, despite
the fact that this ferrite was being operated above its normal frequency range so its loss was unusually high.
There is larger error at small values of the ferrite length when the ferrite is the same length as the coil or
just a little longer. This is because the calculations do not account for the very high radial field at the end of
the ferrite intercepting the coil at these shorter lengths. Although the equations could be extended to cover
this it would be of little practical value because it is easy to avoid this high loss situation by making the
ferrite longer than the coil by 2 ferrite diameters or more.
8.3. Ferrite Loss
The ferrite loss was given by the supplier up to a frequency of 1.5 MHz and was extrapolated to 3.4 MHz
as shown below :
The closest agreement with the data supplied with the ferrite was with Q decreasing as 1/f 1.2, giving an
extrapolated value at 3.4 MHz of Qm = 28. However the theory of domain movement in ferrites which
1
10
100
1000
0.01 0.1 1 10 100
Ferrite Q
Frequency Mhz
Extrapolated Ferrite Q
Suppliers Data
Extrapolated Q
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
010 20 30 40 50 60 70
Resistance ohms
Rod Half Length mm
Series Resistance at 3.4Mhz
Measurements (corrected, see text)
Calculated Ferrite loss resistance
Calculated Total Loss Resistance

Payne : Losses in Ferrite Rod Antennas, Issue 2
11
leads to the loss, gives Q reducing as 1/f (see Hamilton ref 11), and this gives a Q of 35 and this
extrapolation is shown in the curve above and was used in the calculations for Figure 8.2.1.
8.4. Parameter Sensitivity
The sensitivity of the calculated results to errors in the coil parameters was assessed by increasing each by
5% from the values used in Section 8.2 and noting the change in the calculated resistance ΔR, as follows:
Coil diameter : ΔR = +9%
Coil Length : ΔR = -7%
Wire Diameter : ΔR = +4%
So for accurate calculation it is important to measure these as accurately as possible, according to the
definitions given in Section 8.1.
9. MEASUREMENT METHOD
9.1. Correction for SRF
For each measurement the Self Resonant Frequency fr of the coil, or coil + ferrite was measured, and this
was used to correct the measured inductance Lc and the measured resistance Rc using the following
equations (see Welsby ref 5, p 37)
L = Lc [ 1- (f / fr )2] 9.1.1
R = Rc [ 1- (f / fr )2]2 9.1.2
This correction is accepted practice and is based on the assumption that coils have a self capacitance, and
that this increases the apparent resistance and inductance when the coil is used in a series resonant circuit.
However the author has shown that the change in inductance and resistance with frequency is a real change
(Payne ref 14), and so the above factors should therefore be part of the theory, rather than a correction of
the measurements. However for the purpose of comparing measurements with theory it doesn’t matter
particularly whether this ‘correction’ is applied to the theory or the measurements, and since the latter is
accepted practice this was done here.
9.2. Measurement of Inductance and Resistance
Measurements were made with an Array Solutions AIM 4170 analyser, with the impedance of the
connection leads calibrated out. The resistance measurements were subject to large uncertainties because of
the presence of the very high inductive reactance, particularly when the ferrite was present, and so this
reactance was tuned out with a high quality variable capacitor. This had silver plated vanes and wipers, and
ceramic insulation and had a resistance given by (see Payne ref 12) :
Rcap = 0.01 + 800/ (f C2) +0.01 f 0.5 9.2.1
Where C is in pf, and f in MHz
This resistance was used to correct the measured results but it was always very small compared with the
overall measured resistance.
The resistance values were below 5 Ω. The analyser accuracy is guaranteed only down to 1Ω, however a
calibration resistor measured at DC as 2.1 Ω on a meter with a resolution of ±0.05Ω was measured by the
analyser as being between 1.97 and 2.1Ω at frequencies up to 3.4 MHz, implying an accuracy of 1%.

Payne : Losses in Ferrite Rod Antennas, Issue 2
12
10. SUMMARY
Ferrite rod antennas are shown to have a large loss in the wire, in addition to the expected loss in the ferrite.
The author’s equations for the loss in air cored coils is extended to give the loss with the ferr ite added, and
experiment shows good agreement with the theory given.
It is clear that the major loss is in the conductor rather than in the ferrite and so Litz wire should be used
where possible (Payne ref 13). Also the coil should be made shorter than the ferrite by at least two
diameters, to minimise the loss due to the radial flux from the ends of the ferrite.

Payne : Losses in Ferrite Rod Antennas, Issue 2
13
REFERENCES
1. POLYDOROFF W J : ‘High Frequency Magnetic Materials’, John Wiley & Sons, 1960.
2. JOHNSON C & JASIK H : Antenna Engineering Handbook’ 2nd edition 1961, McGraw-Hill Book
Company.
3. SNELLING E C : ‘Soft Ferrites. Properties and Applications’ 1969, Illiffe Books London
4. PAYNE A N : ‘The Inductance of Ferrite Rod Antennas’,
https://www.researchgate.net/publication/351552681_The_Inductance_of_Ferrite_Rod_Antennas
5. WELSBY V G : The Theory and Design of Inductance Coils’, Second edition, 1960, Macdonald,
London,
6. MEDHURST R G : ‘HF Resistance and Self Capacitance of Single Layer Solenoids’ Wireless
Engineer, Feb 1947, Vol 24, p35-80 and March1947 Vol 24, p80-92
7. ARNOLD A H M : ‘The Resistance of Round Wire Single Layer Inductance Coils’ National
Physical Laboratory, Monograph no.9, Sept 1951.
8. PAYNE A N : ‘The HF Resistance of Single Layer Coils’,
https://www.researchgate.net/publication/351373144_THE_HF_RESISTANCE_OF_SINGLE_L
AYER_COILS
9. WHEELER H A : Formulas for the Skin Effect’, Proc. IRE, September 1942 p412-424.
10. GROVER F W : ‘ Inductance Calculations’ Dover Publications Inc, 2004 (first published 1946).
11. HAMILTON N : ‘The Small Signal Frequency Response of Ferrites’
http://highfreqelec.summittechmedia.com/Jun11/HFE0611_Hamilton.pdf
12. PAYNE A N : ‘Measuring the Loss in Air Variable Capacitors’,
https://www.researchgate.net/publication/351413222_MEASURING_THE_LOSS_IN_VARIABL
E_AIR_CAPACITORS
13. PAYNE A N : ‘Litz Cable for HF Single layer Coils’ ,
https://www.researchgate.net/publication/351358011_Litz_Cable_for_HF_Single_Layer_Coils
14. PAYNE A N : ‘A New Theory for the Self Resonance, Inductance and Loss of Single Layer
Coils’ QEX – May/June 2011
Issue 1 : 2014
Issue 2 : July 2021 : Minor corrections
© Alan Payne 2021
Alan Payne asserts the right to be recognized as the author of this work. Enquiries to
paynealpayne@aol.com