The local oscillator in "All American 5ive" vacuum tube
AM radios all drift an annoying amount at the upper
end of the AM BC MW band. The oscillator would be
running at about 2MHz, and warm up drift (from cold
start to about an hour being on) is typically 20KHz.
Enough to make that station at 1520 tune itself out.
AM radios used a hartley style oscillator using the
equivalent of a triode with its plate to B+, grid
capacitivitly coupled to the LC osc tank, and cathode
connected to a secondary winding on the LC osc tank.
Usually an air variable cap, and fixed inductor wound
on a cardboard coil form.

VFO's for ham radio work would involve higher frequencies,
and I would think that they not drift anywhere as bad as
the AM radios did. I looked at a few tube VFO schematics, and
I don't see anything that different from the AM radio
hartley osc circuit. So how did they avoid drift, or were
you expected to leave your VFO on all the time?

The local oscillator in "All American 5ive" vacuum tube
AM radios all drift an annoying amount at the upper
end of the AM BC MW band. The oscillator would be
running at about 2MHz, and warm up drift (from cold
start to about an hour being on) is typically 20KHz.
Enough to make that station at 1520 tune itself out.
AM radios used a hartley style oscillator using the
equivalent of a triode with its plate to B+, grid
capacitivitly coupled to the LC osc tank, and cathode
connected to a secondary winding on the LC osc tank.
Usually an air variable cap, and fixed inductor wound
on a cardboard coil form.

VFO's for ham radio work would involve higher frequencies,
and I would think that they not drift anywhere as bad as
the AM radios did. I looked at a few tube VFO schematics, and
I don't see anything that different from the AM radio
hartley osc circuit. So how did they avoid drift, or were
you expected to leave your VFO on all the time?

Slightly beter components, time to use and change some capacitors to
compensante for the drift, and some stayed on all the time. Voltage
regulation was usually used also.

Voltage regulator tubes, matching components for thermal drift, solid
(physical) construction, for starts. Shielding not only for noise, but
also to minimize thermal drift. If you didn't leave the whole thing on
all the time, maybe you left the filaments on. If you actually turned
the thing off, be ready to wait an hour or so after power up for things
to settle down.

Collins used different techniques -- a permeability tuned oscillator
among them.

Growing up, a bunch of us were hams, living in a few block radius. One
character was really proud of the VFO he'd built; very solid
construction, quality parts, the whole 9 yards.

He had it sitting on a shelf that was mounted to an outside wall of his
house. We'd go pound on the outside of the wall, and you could hear
the thing wobble like mad.


In article , Robert Casey
wrote:

The local oscillator in "All American 5ive" vacuum tube
AM radios all drift an annoying amount at the upper
end of the AM BC MW band. The oscillator would be
running at about 2MHz, and warm up drift (from cold
start to about an hour being on) is typically 20KHz.
Enough to make that station at 1520 tune itself out.
AM radios used a hartley style oscillator using the
equivalent of a triode with its plate to B+, grid
capacitivitly coupled to the LC osc tank, and cathode
connected to a secondary winding on the LC osc tank.
Usually an air variable cap, and fixed inductor wound
on a cardboard coil form.

VFO's for ham radio work would involve higher frequencies,
and I would think that they not drift anywhere as bad as
the AM radios did. I looked at a few tube VFO schematics, and
I don't see anything that different from the AM radio
hartley osc circuit. So how did they avoid drift, or were
you expected to leave your VFO on all the time?

Mostly better design and charericterization. The VFO's designs
were tested for temperature drift, and temperature-compensating
capacitors were made part of the design. I think a few radio VFO's
were actually tweaked as part of the unit test for temperature stability,
but not sure. It would have been expensive.

The capacitors are coded:

N - for negative temperature coefficient
P - for positive temperature coefficient
NP0 - for no temperature coefficient.

So for example, N750 would have -750 ppm of capacitance change
per degree C.

The other challenge was good inductor design, since they can be
temperature dependent as well.

-- Tom



"Robert Casey" wrote in message
...
The local oscillator in "All American 5ive" vacuum tube
AM radios all drift an annoying amount at the upper
end of the AM BC MW band. The oscillator would be
running at about 2MHz, and warm up drift (from cold
start to about an hour being on) is typically 20KHz.
Enough to make that station at 1520 tune itself out.
AM radios used a hartley style oscillator using the
equivalent of a triode with its plate to B+, grid
capacitivitly coupled to the LC osc tank, and cathode
connected to a secondary winding on the LC osc tank.
Usually an air variable cap, and fixed inductor wound
on a cardboard coil form.

VFO's for ham radio work would involve higher frequencies,
and I would think that they not drift anywhere as bad as
the AM radios did. I looked at a few tube VFO schematics, and
I don't see anything that different from the AM radio
hartley osc circuit. So how did they avoid drift, or were
you expected to leave your VFO on all the time?

In article , Robert Casey
writes:

I looked at a few tube VFO schematics, and
I don't see anything that different from the AM radio
hartley osc circuit. So how did they avoid drift, or were
you expected to leave your VFO on all the time?

Several ways:

1) Better components - Drift of the kind being discussed is mostly due to
thermal effects. Capacitors, inductors and even resistors change value when
heated, and the component selection makes a *big* difference in stability. For
example, a variable capacitor with aluminum plates is inherently more affected
by temperature than one of similar construction with brass plates.

2) Better design - Reducing heat reduces thermal drift. High C is usually less
drifty than low C. A high gain tube that is loosely coupled to the tank circuit
is usually more stable than a low gain tube tightly coupled to the tank
circuit. There's lots more, of course.

3) "Weakest link" - As sources of drift are corrected, sources which were once
negligible become dominant. Often a design will go through several revisions as
sources of drift are identified and corrected.

4) Compensation - When all else is done, the use of thermal compensating caps
can reduce drift to very low levels.

Remember too that "stable" is user-defined. A rig that drifts 300 Hz on each
transmission might be considered "very stable" on AM or FM, barely acceptable
on CW, and useless on SSB or FSK/PSK

73 de Jim, N2EY

On Tue, 05 Oct 2004 01:36:33 GMT, Robert Casey
wrote:

The local oscillator in "All American 5ive" vacuum tube
AM radios all drift an annoying amount at the upper
end of the AM BC MW band. The oscillator would be
running at about 2MHz, and warm up drift (from cold
....snippage

I tested an RCA AA5 I have here (uses battery tubes) and
a modified Hallicrafters S120 (also basically AA5) and neither drifted
very far on turn on. The worst was the RCA! at 1.8khz in the first 5
minutes. The RCA drifted 3khz in the first 5 minutes from cold start.
However once both had a chance to fully warm up (20minutes)
the drift was in the less than 100hz region. Generally most AA5
radios for AM broadcast reperesent the lowest possible cost designs
with the least consideration for circuit performance above a minimal
level.

Tubes experence fairly large initial warm up temperature changes
and the surrounding circuits often do as a result of that. When you
consider that your going from 20c to greater than 50C in the first few
minutes there is no surprize there (transistor VFOs would be hard
pressed for that great a temperature change too!).

The solution is manifold. use components that experience minimal
dimensional or other characteristic change for temperature. Coils
and capacitors are the biggest influence here.

The AA5 you have uses a hartly osc and the common circuit has less
than 4 components in the VFO, namely the tube, variable capactor,
a padder cap (vfixed usually) and the coil (slug tuned). Lets look at
each. In the design the tube has changes at warm up of both
mechanical, things move such as cathode to grid spacing when heated
and electrical its operating point shifts as the tube reaches operting
temperture. The mechanical tuning cap, while likely the most stable
device if heated enough the aluminum plates will deform and posible
change spacing, there may be other forces from the chassis mounting
as that warms too. The padder cap is usually a cheap component
in AA5s and the typical part used has a poor temperature coefficienct.
Lastly the coil, this can also be a big factor as the coiled wire can
mechanically change dimension from heating but, you also have a
powered iron or ferrite tuning slug that also has a temperature
coefficient and the cheaper (older) materials can really be poor
with temperature. I might add that some of the AA5s coils were wax
impregnated and the materials used can also experience dimensional
changes while heating up causing the coild to deform. I may add
that operating voltage changes can influence stability and drift.
Tubes are no worse than transistors, just warmer. What differes is
that there are two sources of voltage sensitive drift in tubes, The
heater(filament) voltage must be stable as it affects tube operating
characteristices such as gain and also the environment due to heating
of the area around the tube. The other votage that must be stable is
the B+ (high votage usually but can be anywhere from 12 to 300v
depending on tube and circuit).

Hopefully you can see that VFO design superficially can be very simple
but has many details that can influence stability. It is possible to
design a tube (or transistor) VFO that is very stable but it requires
good components, temperature compensation and good mechanical
construction.

If you want stability in your AA5... Better cooling most ventilated
very poorly. Other things, put the components away from the tube
[but not too far ] to minimize heating effects. Regulate the
voltages. The latter is harder as the average AA5 uses a series
heater string that has a sometime shakey warmup. Also the AA5
uses a poor supply in the form of a 35w4 half wave rectifier that has
to warm up to work (it's a tube too!) and the lack of good power
supply filtering (add to this the caps are old!!).

As a experiment with tubes:
The Hallicrafters was a basket case when I got it so modiflying it was
a reasonable thing to do (just to make it work) lest the purist
classic radio people protest. First was converting from live chassis
(typical AA5 off the mains) by adding a transformer to supply 120V for
rectifier and 6.3v for the heaters. Rewired the heaters for 6.3V and
change the 50C5 to a 6AQ5 I had on hand (rewire socket). This radio
used a selenium rectifier which was bad so silicon bridge rectifier
was used with new 100uf caps as filter to get clean 150v B+. Since
If selectivity was minimal I added an old 6KHZ mechanical filter and a
second IF amp tube (5899 subsub mini) to fix that and and a bit of
gain. Replace a dozen poor quality and just plain bad caps in various
points including the Local osc section (VFO). I also added a 6AR5
bfo/product detector for ssb operation as the original BFO design was
poor at best. It was a major rebuild with many circuit changes mostly
for fun. However the 6BE6 and the Hartly local osc (VFO) was
retained to keep the tuning dial something near calibrated. A stable
VFO using tubes was straghtforward using good quality parts ( both the
BFO and the LO are LC osc). The result is a stable (after warm up)
general purpose reciever that I use for 3885KHz AM and occasional 75m
SSB listening.


Allison
KB!gmx

Actually the drift is thermal not due to voltage variations. Voltage
regulation is not an issue these days with the power line usually holding to
better than one percent.

In order to halt the drift, you need to replace some of the tuning
capacitance with negatifve temperature coefficient capacitors. How much you
use will depend on how much the drift is for a given temperature variation.
It drifts more at the high end because there is less capacitance involved,
causing a small variation to make the frequency move farther.

You can start with a negative temperature coefficient capacitor of, say 10
pF across the tuning capacitor and see if you can still align the local
oscillator. Mount the capacitor as close to the source of heat (the tube)
as possible and see what happens.

I have compensated many oscillators for thermal drift. You can also see if
perhaps the existing components are being heated by a power resistor and if
possible increase the distance. There is a bit of art involved but it's all
pretty basic.

The big problem, is where can you buy negative temperature coefficient
capacitors? They used to be plentiful but I don't know about these days.

Bob

Hi,

At the high end of the AM band, the oscillator capacitor is
at its lowest capacitance (almost fully un-meshed). So, any
variations in circuit capacitance due to tube electrode
dimensions, capacitors with non-zero tempcos or similar
happenings in the oscillator coil will have a greater effect than
at the low frequency end.

For this reason, a stable LC-tunable VFO would not normally
be designed to tune the 3:1 ratio that an AM local oscillator is
expected to.


Cheers - Joe

VFO's for ham radio work would involve higher frequencies,
and I would think that they not drift anywhere as bad as
the AM radios did. I looked at a few tube VFO schematics, and
I don't see anything that different from the AM radio
hartley osc circuit. So how did they avoid drift, or were
you expected to leave your VFO on all the time?

Some VFO's e.g. in the BC221 use a compensation
coil, moved by a bi-metal, inside the tuning coil as temperature
compensation.

Wim

Joe McElvenney wrote in message ...
Hi,

At the high end of the AM band, the oscillator capacitor is
at its lowest capacitance (almost fully un-meshed). So, any
variations in circuit capacitance due to tube electrode
dimensions, capacitors with non-zero tempcos or similar
happenings in the oscillator coil will have a greater effect than
at the low frequency end.

For this reason, a stable LC-tunable VFO would not normally
be designed to tune the 3:1 ratio that an AM local oscillator is
expected to.


Cheers - Joe

Joe, back in the 'old days' everyone left his VFO energized on a
24-hours basis, to improve stability. Same with our receivers,
particularly when SSB became the vogue.

Harry C.