88 mH Telephohne Loading Coil
Telephone Induction Coil
one of the basic electronic passive components along with capacitors
A capacitor stores energy in a electric field and an inductor
stores energy in a magnetic field. The capacity of a
capacitor can be increased by using a dielectric material
between its plates and an inductor can have its inductance
increased by using a magnetic material as its core.
was getting started resistors were made by winding a specified
length of wire on a bobbin. When Tesla was working in
Colorado Springs it was common to specify the inductance of a
coil in meters, and the same unit was used for
capacitors. Inductors were commonly used in tube type
electronic equipment and there were a large number of them
available as component parts. But they typically weigh
more, take up more volume and cost more than capacitors or
It's very difficult to include an inductor in a semiconductor
type integrated circuit (IC). Prior to ICs to make a low
pass filter you would use series inductors and shunt
capacitors, but now it can be done with an op amp and just
resistors and capacitors. In fact all the common filter
types can be realized without using any inductors. With
the advent of Switching Mode Power Supplies (SMPS) inductors
are much more popular. These require the ability to
handle the currents involved in a specific design and be small
and low cost. The efficiency of the SMPS depends in a
large measure on how low a resistance the inductor exhibits
and in fact the specifications for the IC depend on the
quality of the inductor used. so it's common that the
evaluation boards for a SMPS will use an exotic inductor.
Joseph Henry (Wiki
was the first person to make a close wound electromagnet
that was used to mechanically ring bells, like to call a
servant. He used his wife's spinning wheel to wrap silk
around the wire to form insulation (if you have details on how
this was done, contact me
Prior to this all the work done with electromagnetism was done
with bare wire would with a large air gap between turns, or a
single turn. Insulated wire led to the Telegraph
which in turn led to Teletype machines that use
a binary code for the characters. My first computer
used a Teletype model 33 Automatic Send Receive (ASR) for
keyboard input, printing, paper tape punching and paper tape
By adding an insulation that was very thin Henry was able
to close wind his wire on a soft iron "U" core to make a
lifting magnet that could hold many hundred pounds.
All the prior experiments with electromagnetism were done
with bare wire and so Henry was the first to demonstrate
"Soft Iron" (Wiki)
was the first core material. The soft has two
meanings. First, like bailing wire (Wiki)
and floral design wire it is easy to bend and shape.
Second it does not get magnetized so when the electric
current is turned off the electromagnet has almost no
magnetic effect. Soft iron can support over a 2
Tesla without saturating and so is commonly used for
electromagnet pole pieces.
In the early days solid soft iron cores were used in
and solenoids (Wiki).
Ignition Coils (Wiki),
Faradic (quack medical) devices,Violet Ray (quack medical
devices), early telephone
induction coils used in phones all used a core made
up of a bundle of fine soft iron wires.
Later, in some applications, Electrical Steel (Wiki)
replaced soft iron for applications where energy
efficiency mattered. For example in self
winding (Western Union) clocks the clocks made in
the late 1800s uses solid soft iron cores, but the later
clocks used laminated Electrical Steel for the
Electrical Steel laminations are used as the core of AC
transformers. Not only does it improve the
efficiency it also allows a construction method where the
laminations are built up in overlapping layers.
and Powered Iron (Wiki)
cores are used to make inductors. These may be in
the form of a toroid (Wiki)
where there's a closed magnetic path and so very low
magnetic leakage or susceptibility. These cores are
also available with an air gap to prevent saturation where
the application needs that.
When the first turn is would on a core the filling factor
comes very close to being 1 (it's a little less because
the wire insulation and maybe there's paint on the
core). But when a second winding is added on top of
the first winding the copper wire of the first winding
takes up some of the area under the second winding
reducing its fill factor. As more and more layers
are added they have less and less inductance per turn.
Faraday's law of induction (Wiki
- The induced electromotive force (EMF) in any closed circuit
is equal to the time rate of change of the magnetic flux
through the circuit.
Ampère's circuital law (Wiki
- relates the integrated magnetic field around a closed loop
to the electric current passing through the loop.
Lenz's Law (Wiki
An induced current is always in such a direction as to oppose
the motion or change causing it
To get a high Q in operation the load capacitance needs to
be much higher than the self capacitance of the coil.
That's so say the coil need to be operated at a frequency
that's much lower than it's self resonate frequency.
To lower the self capacity of a coil you need to pay
attention to both the turn to turn capacitance AND to the
voltage between those turns. For example:
1. In a single layer cylindrical coil with a lead at
each end and it's tight would you can approximate the
capacitance between each turn by assuming the turns are
insulated rings. BUT, you also need to look at the
voltage between the turns. In this case the voltage
between adjacent turns is Vtotal/N where N is the number of
2. In a double layer tight wound coil where the input
and output wires are adjacent to each other you can use the
same ring approximation for the capacitance between adjacent
turns. But now there's a huge problem because of the
voltage between turn 1 and turn N. Between these two
turns the voltage is Vtotal. That's N times higher
than in the case above. This is why using bank winding
(see the Loop antenna page) is
used as well as breaking the coil into sections is used to
minimize the voltage between turns.
See my Q Meters web page for
The best reference book is Tesla's The Colorado
Springs Notebook (ISBN-13:
978-9562914628). In the early part he did not have
high Q operation and showed photos of each complete
setup. Later, after adding a metal pipe vertical
antenna with capacitive top loading so that the secondary
coil was operating with a load capacitance much higher than
it's self capacitance, he only photographed parts of the
setup to keep some secrets. He mistakenly thought he
had power multiplication, rather than just the voltage
multiplication that comes with high Q operation.
Note: the demonstration "Tesla Coils" that are sold for
school demonstrations are not really high Q coils but rather
are just RF transformers. Their output voltage is just
the input voltage stepped up by N^2. In a true Tesla
coil there would be a capacitor in parallel with the
secondary and the output voltage would be Q * N^2.
Note it takes measuring the spark length between a couple of
large metal spheres or a high voltage probe (these are over
$1,500 used on eBay) to measure the voltage, not the gap
The basic unit of inductance is
the Henry (Wiki
named for Joseph
(Wiki) who in 1828 made the first practical
electromagnet (see my electronic time line
Any coil of wire will have some resistance and an ideal
inductor would have zero resistance. There are a number
of sources of resistance and the show up a different slopes is
a plot of coil Q vs. frequency. The DC resistance does
not change with frequency. The resistance because of
skin effect (Wiki
and proximity effect (Wiki
can be mitigated by using Litzendraht wire (Wiki
The dielectric loss (Wiki
in the wire insulation is also a factor. For example some
insulation materials are lossy at RF frequencies as some coil
coatings may be.
There is also capacitance between the turns of an inductor and
as the frequency is swept the impedance (Wiki
page) of the inductor goes
through at least one resonance. At a fixed frequency the
ratio of the inductive reactance (Wiki
to the resistance is called the Quality or Q (Wiki
the inductor. When used to form a resonant circuit with
an external capacitor the Q of the circuit is inversely
proportional to the bandwidth of the circuit. Varnishing
or coating a coil to hold the turns in place or moisture proof
the coil also adds capacitance.
Most inductors have terminals
rather than coax connectors and so require some sort of test
fixture. The test fixture has parasitics that need to be
backed out of any measurements.
There are a number of ways of measuring inductance. One
way of characterizing the different methods is to look at the
measurement error as a function of frequency. This is
covered in The Impedance Measurement
by Agilent. The most accurate method
makes use of the 4-terminal pair connection method as used by
such instruments as the HP
4274 and HP 4275 LCR Meters
For measuring the Q specialized equipment is more accurate
than using general purpose equipment. The Boonton 160 Q meter
was an early instrument that
uses a 4-pin vacuum tube. There was a motorized
accessory for the HP 4284A that allowed it to measure Q.
Method and Apparatus for Electrical Measurements, H.A. Snow
(Boonton Radio Corp), Nov 22, 1938, 324/654; 324/653; 338/61 -
324 Electricity: Measuring and Testing
653 Impedance, Admittance or other
Quantities Representative of Electrical Stimulus/Response
Lumped type parameters
For figure of merit or Q value
654 Impedance, Admittance or other
Quantities Representative of Electrical Stimulus/Response
Lumped type parameters
Using inductive type measurement
338 Electrical Resistors
61 with Inductance-Reducing
The HP 4395A can measure the
of a series connected component over the
frequency range of 10 Hz to 500 MHz with a Resolution Band
Width (RBW) as narrow as 1 Hz.
Rather than looking at the inductance and resistance at a
single frequency a better method is to make a swept
measurement and fit a model to the data. This way you
capture not just a 2-element model, but a more complex
model. This is the professional way to measure crystals.
June 2013 - as part of my work
on the Joule Thief
looking into ways to identify toroid cores that are unknown
manufacturer and material.
Note the Tnn size designation is based on the OD in inches.
A T100 core would have an OD of approximately 1 inch, a
T 50, half inch, etc.
The first step was to make a spreadsheet listing the sizes
down the left side and the materials along to top with AL
values in the main body.
Note ferrite toroids are commonly named FTnn, but I've dropped
the F and list them with all the other toroids.
A blank cell or "na" in a cell means no part with that
material and size was found on the first try.
There were a number of sizes where there was a suffix letter
like T50, T50A, T50B, T50C & T50D.
If a high letter suffix was found but no lower letters they
were listed anyway.
In the case of T200A and T200B they seem to be the same?
It takes 6 letter size pages to print, so here's a photo of
the AL chart.
It seems that the first step in figuring out what material you
have is to measure the physical OD, ID and Height of the core.
This will eliminate the vast majority of the possible cores.
Now measuring the AL value should narrow down the possible
Example No. 1:
Marking: 8620 55305-A2
The T90 is: OD 0.900, ID: 0.55, Ht: 0.3 and the T90E is OD:
0.90, ID: 0.55, Ht: 0.3
Don't have dimensions for the T90A, T90B, T90C or T90D so not
sure what size this is.
Example No. 2:
Color: body: light green, one face: Blue
The ID is slightly smaller than the T130 listed & the Ht
is a little bigger.
Example No. 3
Marking: 482S 77934-A7
Color: shiny black
This is very close to the T106 and so might be material: 60,
63, 66, 70, M125, 26, 52, 1, 2, 3, 6, 7, 12, 15, 17, 0
An AL measurement would narrow that down. It's also not
a ferrite, but rather powdered iron.
Example No. 4
Color: light green body, one face blue (same as Ex No. 2
This is a T200A (same as T200B) so the material might be: 26,
52, 2, 60, 66 all powdered iron
Example No. 5
Color: light green/blue one face yellow
This is a T300D and the material might be: 60, 63, 66 oin
Example No. 6
Marking: -MB4- 55321-04
Color: light brown
Example No. 7
Marking: 563S 77548-A7
Color: shiny black
OD: 1.3" (broken so a guess)
ID: 0.82 (broken so a guess)
By comparing this one with Ex. No. 3 the marking might be
A toroid is shaped like a doughnut. Since it has a
closed magnetic path there is little magnetic leakage if the
turns are a single layer wound close to the core.
Different materials behave differently with frequency and
These toroids were ordered for experimenting with the Joule Thief
These are Fair-Rite
RFI suppression cores.
Distributors that carry toroids: Amidon
also see my Disty
There are different ways a toroid can be wound.
One way it to get as many turns as possible. For these
applications the "Fill Factor" becomes important.
What you're looking at is how many square mills of wire area
can you fit into the hole area. If the wire diameter
was very small the fill factor will be near maximum but as
the wire size increases the fill factor gets smaller because
of the air gaps were there are no turns.
Another way is to wind a single layer coil. Again
smaller wire sizes allows more turns than larger wire sizes,
but smaller wires have more loss.
Length of a Turn
For a single turn of a single layer wound toroid:
Lturn = 2*(OD-ID) + 2*Len
Example for the first two rows of the table above, Lturn =
(0.38 - 0.197) + 2 * 0.19 = 0.563
Then 10 turns would take 5.63" of wire plus a couple of 4"
leads is about 10" of wire.
Each toroid has an AL
value that allows the number
of turns needed for a given inductance to be estimated.
N = SQRT(L/AL
AL = L / (N*N)
Examples with 10 turns, AL
= L(10 turns) / 100
First row AL
= 50uH /100 = 0.5 uH
Second row AL
= 170uH / 100 = 1.7 uH
For inductors that have magnetic
cores (typically without an air gap) there is a maximum limit
to the amount of magnetic flux that can flow. At some
point the core saturates and no longer has magnetic
properties, i.e. it's as if the core was removed and replaced
by air. At very low flux the permeability appears to be
a constant value, but as the flux approaches the saturation
value the permeability starts to decrease (approaches 1 = free
space or air).
In a magnetometer (Flux Gate Patents
when the core is saturated there's no change in the Earth's
magnetic files, but when it's not saturated the Earth's filed
is concentrated. The difference between these causes an
AC output proportional to the field strength (especially at
the second harmonic of the drive frequency).
The core memory (Wiki
used in early computers requires that the core remains
saturated in one of two directions. The X and Y drive wires
carry just over half of the needed amp turns to flip the core
magnetization so only the core that has both X and Y driven
will be flipped (if it was magnetized in the opposite
The energy stored in an inductor
is given by E = ½ * L * I2
But the current (I) is limited by saturation of the core
material. A way around that limitation is to insert a
narrow air gap into the core. When measuring the core
current in the Joule Thie
another core that was the same as the working core was used as
the current transformer and the drive current was limited by
the source impedance of the HP 33120
Function Generator. A more flexible method would be to
use a FET and bench power supply so that the source resistance
would be near zero and the voltage could be higher allowing
for higher currents.
The energy is stored in the non magnetic parts of the core
structure, either in an air gap or in the non magnetic binder
that holds the material together.
For a transformer where you want low core losses it's good to
have a lot of magnetic material and not much air.
Electrical steel (Wiki
is used in thin sheets to minimize Eddy current (Wiki
losses. This was a big improvement from solid iron bar
cores. The use of a bundle of soft iron wires in the
core of induction coils (Faradic
, quack UV light
) was an improvement on the solid iron
rod cores of the first generation coils.
88 mH Telephone Loading Coil (Wiki)
In about 1887 Oliver Heaviside (Wiki)
knew that if R/G = L/C there will be minimal distortion in a
transmission line. But most phone and telegraph lines
were operated such that R/G >> L/C. In 1893 he
proposed adding discrete inductors at periodic intervals along
the line. He was working for AT&T at the time and
they started a patent application, but Putin beat them to it
with less understanding of what was going on.
Mihajlo Pupin (Wiki
patented Pupin Coils in 1899.
Trivia: It's not unknown for the inventor to not
understand how his device really works. An example is
the Griffin Grinding Mill (see my Gyroscope
Loading coils come in two different colored wires (red and
green). A single winding has a DC resistance of 4
In the telephone application these were used to "load the
line". There are two key ideas: 1) If the signal could
be transformed to a higher impedance the line resistance would
cause lower loss. This is the method used to deliver AC
mains power by means of transformers. Transformers are
not a viable option in telephone circuits because they do not
pass the DC signaling, but the line impedance can be
raised using series inductors. There is a decrease in
line loss of about 1.4 dB per mile (1.6 dB/mi to 0.2
dB/mi. It's best if a 500
telephone is used that compensates for the distance between
the phone and the central office. Older phones
will not work as well. 2) A low pass filter can be
formed where the inductance is that of the line plus the added
loading inductor and the capacitance is that of the
line. For 88 mH coils the typical distance between coils
is 6,000 feet. At the start and end of the line the
distance is 3000 feet. Testing should be done within
that range and is typically done at 1 kHz for audio signals.
Also see my web page on line impedance
when operating below the critical frequency. Details on
the use of loading coils can be found in BC Burden's 1948
Handbook for Telephone Managers and Engineers.
Note: Loading should only be done between the central
office and a single phone. If a party line phone is
connected somewhere in the middle of the line the loading will
not work properly and the loss may (probably will)
increase. A party line could be implemented by having
home run circuits from the CO to each subscriber and bridging
them in the CO using bridge lifters (Saturateable inductors
that present a high impedance at voice frequencies to the
line(s) that are not in use to prevent them from imposing a
shunt capacitance across the line circuit, which would degrade
transmission on the pair that is in use. The DC current
through the inductor connected into the active line saturates
it, causing its impedance to drop to a negligible value.)
Wiki: Primary line constants: Loaded Line
Some vacuum tube power supplies included a "swinging choke"
whose inductance changed with the current flowing, i.e. very
similar to the telephone saturable inductor.
|As received the same
color coils were connected together by adhesive
tape. Some care was used when separating each
|There are four leads
since this is a dual coil, or transformer. Just as the
schematic symbol has one of the wires for each winding
marked with a dot this has one wire marked with yellow
When testing a winding (one yellow & one non yellow
found by experiment) it measures about 22 mH and a Q of
But when connected as shown (in series) the inductance
is 87 mH with a Q of about 200. Note doubling the
number of turns should multiply the inductance by 4
since inductance in proportional to (# turns)^2.
It's my understanding that when used in phone service the
individual coils would be connected in series with the tip and
ring wire pair. Some type of test would need to be done
(you can tell from the mechanical layout of the coil) to be sure
tip went to tip, i.e. that the wires were not reversed.
They are connected so the yellow sleeve points the same way for
the DC current, i.e. one pointing to the CO and one pointing to
Apparatus for Telegraphic or Telephonic Transmission, M.I.
, May 8 1894, 381/98; 307/102; 379/41 -
of Reducing Attenuation of Electrical Waves, M.I.
, June 19 1900, 178/45 -
Magnetic core for inductance-coils, John C
App: 1901-11-30, 178/46; 148/277; 336/210; 336/213; 29/607;
252/62.55; 336/234; 29/605 -Ferrite core
Loaded electric circuit, John C
App: 1901-11-30, 178/45; 336/213; 336/234; 178/46; 336/229
- improvement over solid iron or laminated iron cores.
Loaded phantom-circuit, George
1911-01-10, 370/200; 178/45; 178/46
1916-03-07, 178/46; 370/200; 370/201
phantom loading coils
, 1932-04-12, 336/234; 148/310; 148/311; 148/312; 420/459; 420/581
"repeating coil with primary and secondary, the core of the coil
consisting of alternate layers or laminations of silicon steel
and high permeability nickel-iron alloy commonly known as
permalloy,...". works over a wider range of line voltages
than prior coils.
- has info on various model numbers and sizes.
Also see my Telephone
& telephone patents
Telephone Induction Coil
These are used as part of a telephone set either as a
transformer or as an inductor.
The construction is similar to the violet ray quack medical
device and I suspect other "induction" coils in that their
magnetic cores are made up of many individual wires or in
the case of two of these No. 46 induction coils thin iron
plates. Using laminations
is another way to make more efficient magnetic paths, but
things like "E" and "I" or "L" and "I" shaped laminations
are not appropriate for this application (these have a part
of the path in air so they will not saturate). This is
an improvement on the solid iron cores used in early telegraph equipment.
Resistance: 1-2: 16.48 Ohms, 3-4: 26.78 Ohms.
Note coil acts as an electromagnet since the core is not
closed. The core is made up of straight soft iron and
works at least through 3 kHz voice frequency.
With a DC Gaussmeter touching the end of the core the
magnetic field in Gauss is about the same as the current
through the 1-2 coil in mA.
Telephone, T.A. Edison, 1896-11-25, 379/387.01; 192/84.1; 24/132R -
oldest patent in class 379/387.01-
I is the Induction (inductorium) Coil
Means for utilizing secondary batteries on
telephone-circuits, C.E.Buell, 1882-06-20, 379/387.01 -
|SR 715 LCR
|88 mH (connected)
Telephone Loading Coil (Red)
|4 + 4
65 (1 kHz)
|88 mH (connected)
Telephone Loading Coil (Green)
|4 + 4
65 (1 kHz)
|Three No. 46 Telephone
Induction coils (transformers)
Note 1: the Fly Back (Ring Tester) LEDs are codes as 1R, 2R, 3R,
4R, 1G, 2G, 3G, 4G with 1Red being the poorest result and 4G
being the best result.
Method of forming coils for electrical apparatus, Edward
L Aiken, GE,
1903-11-17, 72/371; 72/137; 29/609 -
E Hadley, Federal
Manufacturing Co, 1905-07-25, 242/483.4; 74/59;
Electrical coil and method of winding same, Max
Helm, 1913-12-09, 336/206;
206/389; 242/166; 242/178; 336/209; 242/444.4- basket
weave for magnet and resistance coils (not inductors).
Ryden, Universal Winding Co, 1915-06-01, 242/480.8; 242/482.7- for wire coils
Electrical coil and method of winding same, Claes
Ryden, Universal Winding Co, 1916-11-07, 336/206; 242/166; 242/178; 336/209;
242/444; 242/447.1 -
A Freeburg, 1929-01-22, 336/191; 29/605;
336/107; 336/67; 336/208; 336/229; 336/222 -
Coil winding machine, W.O. Meissner, 1931-05-26, 242/444.5 - multiple parallel coils, fine
lacquered wire, wax paper layers
Coil winding machine, Hoxsie
W Lillibridge, Atwater
Kent, 1931-06-30, 242/447.2 -
Radio-amplifying circuits, Wladimir
J Polydoroff, Johnson
Laboratories, 1933-12-19, 330/169;
330/155; 334/77; 336/206; 336/234; 330/171; 334/61; 336/136;
336/226; 361/298.1 - powdered iron cores
Magnetic core material, Wladimir
J Polydoroff, Johnson
Laboratories, 1934-12-04, 336/233;
148/306; 29/608; 106/228; 106/287.18; 252/62.54; 264/325;
428/900; 75/348; 106/253; 106/287.23; 148/104; 252/62.53;
264/DIG.58; 428/402- torodial & pot cores
Variable inductance device, Wladimir
J Polydoroff, Johnson
Laboratories, 1934-10-30, 336/136 -
movable core of insulated magnetic particles
Permeability-tuned resonant circuit, Wladimir
J Polydoroff, Johnson
Laboratories, 1934-10-30, 334/65; 334/75;
334/77; 334/85; 336/87; 336/136; 336/208; 334/76; 334/80;
336/67; 336/90; 336/205; 336/231 -
Coil winding machine, Robert
H Burns, RCA,
1935-09-17, 72/133; 72/142 - Vacuum tube
Magnetic core for high frequency inductances, Wladimir
J Polydoroff, Johnson
Laboratories, 1936-11-03, 331/176;
336/155; 336/182; 336/233; 331/167; 336/179; 336/212;
Loop antenna, Lewis
H Van Billiard, GE,
1942-08-04, 343/866; 29/605; 336/205; 336/206;
343/702; 343/842; 174/396 - shielded with deep nulls
Machine for winding electrical coils, Frederick
N Jacob, Martin
J Kirk, Johnson
Laboratories, 1942-12-15, 242/447.1; 388/824
- cylindrical form, basket weave
Back to Brooke's Home
, Products for Sale
Page created 19 Jan 2011.