Modern Methods of Obtaining Bias
from Mallory-Yaxley Radio Service Encyclopedia, April, 1938
Modern Methods of Obtaining Bias
Mallory Bias Cells
(Mallory-Yaxley Radio Service Encyclopedia,
April, 1938, pp. 214-215)
One of the
most perplexing problems faced by the radio design engineer has
been the method of obtaining bias for various tubes. Unless
large by-pass condensers are used with a filter network, cathode
bias, or bias derived from the voltage drop in the negative lead
of the power supply system, has many disadvantages. Common
impedance, and voltage fluctuations with signal in the power supply
circuit causes a continuous variation of bias voltage. The
bias voltage variation is responsible for many undesirable effects,
such as instability, loss of bass note amplification, and distortion.
As previously explained, the by-passing efficiency of a
condenser varies with the frequency. At the low frequencies
where the by-passing action is most needed, the efficiency of
the condensers is the lowest. Condensers of sufficient size
to prevent objectionable bias variation are both bulky and expensive.
Cathode bias has the further disadvantage of placing the
cathode at a different potential from the heater, so that electronic
emission from the heater will frequently cause an objectionable
amount of hum.
In the past the
only alternative methods of obtaining bias were through the use
of a separate "C" bias power supply, requiring a separate
rectifier tube, choke and filter condensers, or by using a common
"C" battery. It will be generally agreed that
a "C" battery has little place in a modern AC receiver,
transmitter, or PA amplifier as such batteries have a tendency
to become noisy and develop high resistance with age, and frequently
fail when most needed
Stated briefly what
was needed was an independent source of "C" bias that
was inexpensive, compact, absolutely quiet, and having a life
equal to the other components of the circuit - the tubes, resistors,
and condensers. It was unnecessary for the bias supply to
furnish current or power because no appreciable grid current flows
in an amplifier tube operated under Class "A" operating
After many years
of research the problem was successfully solved by the design
of the Mallory Grid Bias Cell. Over three million Mallory
Bias Cells are now in use. They are standard original equipment
in most modern receivers.
WHAT IS A BIAS CELL?
CHARACTERISTICS - Physically,
the bias cell is a small acorn-shaped device 5/8" in diameter
and 11/32" deep. The black disc is the positive electrode,
the metal container is the negative electrode.
VOLTAGE - Bias cells are produced in two types: the
standard 1-volt cell, and the new 1-1/4-volt cell. The two
types may be distinguished by the fact that the 1-1/4-volt cell
has indentations or concave depressions in the electrodes, while
the 1-volt type is smooth.
For experimental applications the choice between the two types
of cells should be made on the basis of the voltage that is required.
When making repairs on commercial receivers, replace with
the type of cell that was used as original equipment.
CURRENT (DC) - The cell is strictly a potential, or
voltage, cell and should not be used in circuits in which currents
of over one microampere or .000001 ampere is drawn from the cell.
CURRENT (AC) - The characteristics of the cell are unaffected
by superimposed AC as high as 360 microamperes of any frequency.
TEMPERATURE - The cells may be used in ambient temperatures
from 40º below zero to 120ºF. The voltage of the
cell remains reasonably constant throughout this wide temperature
range. It is recommended, however, that wherever possible
the bias cells be placed in the coolest location.
HUMIDITY - The cell exhibits no change in characteristics
when exposed to a relative humidity of 90% at 120ºF.
IMPEDANCE - The bias cell is non-reactive at audio frequencies
and the DC resistance ranges between 11,000 and 50,000 ohms.
NOISE - The cell does not cause the development of any
Under no circumstances
should the Mallory Grid Bias Cell be checked with an ordinary
voltmeter, regardless of the sensitivity of the meter (ohms-per-volt
rating). If for any reason it is desired to read the voltage
of the cell, a vacuum tube voltmeter (VTVM) should be used which
has at least a 10 megohms input impedance.
To insure complete satisfaction from the use of the Mallory
Bias Cells, the following simple precautions should be observed:
1. Bias cells should not be used to bias oscillators,
power output tubes, or amplifier tubes in which grid current may
2. The new 1-1/4-volt Mallory Bias Cells may be
mounted in any position. The standard Mallory 1-volt bias
cells may be mounted in any position except with the black electrode
in a horizontal plane at the top of the cell.
3. To avoid accidental short-circuiting of the
cell, keep it in its envelope. Do not permit cells to be
carried loose in a pocket with metal objects such as coins and
4. When soldering bakelite insulated holders use
care not to over-heat connections or loose terminals will result.
5. Many of the new model receivers and amplifiers
are using 1-1/4-volt Mallory Bias Cells, mounted in wire spring-type
clips. Should it be desired to replace these cells, care
should be taken not to deform the clip, or bend it so far that
inadequate contact pressure will be maintained on the electrodes.
The wire clips can best be spread by using a special tool
that is made for the purpose; lacking this, a pair of long-nose
pliers can be employed if the outer faces of the jaws are notched
slightly to prevent slipping.
Bias Cell Rejuvination
Posted 12-15-98 to Deja News and rec.antiques.radio+phono
I assume that the cells are the
standard 'Mallory' style which looks like an upside down classic
flying saucer. Examine the flat side which is a flat carbon disk
and see if the rubber seal/insulator is still intact. It may be
a bit perished around the opening, from exposure to the air etc.
but it must be in fair condition where it is between the zinc
case and the carbon disk. If so, you will need a #60 to #65 drill
bit, an insulin syringe and a dab of clear RTV sealer. With the
drill held in a holder like a small exacto knife handle (so you
have some `feel' for what's happening) carefully drill down through
the carbon near the edge heading for the cup in the zinc. When
you feel the drill break through the carbon disk, STOP!!!! Below
the carbon is an insulating separator and below that ???
Fill the syringe with some distilled
water and poke the needle carefully through the hole and the separator
(it's soft like blotting paper or fabric) and inject a few drops
of water below the separator. If water instantly comes out of
the hole you are not into the space in the back. STOP. Go deeper.
DO NOT inject too much water, we don't want it to start to come
out of the hole since it will then be shorting from the back (-)
to the carbon (+).
Once you have injected the few
drops, wait a few minutes for it to soak in and then check from
the zinc - to the carbon + with a VTVM or DVM and if all went
well, the cell should have about 1.2 to 1.45 V showing. If it
is now OK, then, with a tooth pick, apply a tiny spot of RTV into
the hole to seal it. Keep the RTV off the carbon face, so it does
not insulate the contact finger of the holder. If you still have
0 V then you will need to make an adaptor to install a silver
watch cell in place of the old one. Remember the silver cell must
go in face down so the + is outward, and the edge of the cell
must be insulated from the holder.
Alternate Method of Water Injection by Dave Froehlich, from emails.
If the carbon and zinc are still good, when water is added to the dried minerals,
which were the original electrolyte, the battery will work again. But
there's no need to drill a hole. Zinc is a very soft material and it
crimps the carbon cover in place. So all that needs to be done is
carefully pull up the edges around the carbon disc and it comes right
out. Then add water, and put the disc back on and crimp it in place.
There is already an insulator around the edge of the disc. That's it.
The cell is back to 1 volt. I just did it. There is no odd material
under the disc. That's the dried out electrolyte that will get mixed
with the new water. I would also like to mention that the crimp around the carbon button may not
look like new after re-crimping, unless someone is good at metal sculpture.
If you want to presserve the original appearance of the cell, you may have to go with
the hole method. But the way I describe gets water where it needs to be in a hurry.
PHILCO MODELS USE GRID-BIAS CELLS
from Philco Serviceman, February, 1936
The new Models 602, AC/DC receiver, and 624, six-volt battery receiver, both employ the new grid-bias cells which have recently been introduced in commercial receivers. These cells are small, acorn-shaped units, which are held in a special socket in the receiver. In the Model 602 there is one cell employed, and in the Model 624 three of them are used. The case is the negative electrode, and the disc is the positive. The no-current potential of the cell is one volt. Its life is practically indefinite.
There are certain disadvantages in some types of circuits in using self-bias on detector and audio tubes. These new grid-bias cells overcome the disadvantages and afford much better circuit characteristics and minimize the possibility of overload distortion. The bias cells are connected in the grid circuit in all cases. The negative (case) is connected to the grid end, and the positive (disc) is connected to the ground end of the circuit. In the Model 602 the one-megohm audio filter resistor is connected in series with the bias cell, and the entire circuit is then bypassed to ground with a .1 mfd. condenser. In the Model 624 three cells are used in series, these being connected in the grid circuits of the type 32 first audio and the type 30 driver tube. A .05 mfd. condenser is connected from the negative point to ground.
These cells are not adapted for current and should never be tested with a voltmeter. The only way that the voltage can be checked is by means of a vacuum tube voltmeter. Since most servicemen do not have this type of equipment, the only thing to do is change the cell when it has become damaged for some reason.
The new Philco grid-bias cells open a number of possibilities to experimenters. The individual bias cell is known as Philco Part No. 41-8009 and sells at a list price of 20 cents. The bias-cell panel assembly, consisting of three sockets for the Model 624, is Philco Part No. 38-7275 and sells at a list price of 20 cents. The individual socket, as used in the Model 602, is known as Philco Part No. 38-7436, list price 15 cents.
BACK TO TOP