|
Two things always
bothered me about my
E F Johnson Viking Valiant and
E.F. Johnson Ranger
transmitters:
- When I would
put the VFO on
spot and listen to
the beat note on
higher bands, the
beat frequency
changed
significantly as I
activated and
tuned the power
amplifier.
- Sometimes I
would get
incurable chirp
and frequency
jumping on higher
bands, generally
worse when the VFO
was running
straight through
on 40 meters (later stages not
multiplying).
Finding the
problem was a fairly
difficult job. I
tried resistor
loading the VFO
output to further buffer
the VFO,
I tried adding voltage
regulation to the
plate supply of the
VFO tube, I even
added an
additional buffer
circuit, and added
extra RF
bypassing. I tried
cleaning the panel
bushings on the VFO.
I finally figured
out the problem was
ground currents
looping through the
shafts and front
panel, and getting
into the VFO through
the VFO shaft and
other housing shield
gaps.
See how a shield
works at this link.
Problem Cause
E.F. Johnson, in
many rigs including
the Ranger 1, Ranger
II, and Valiant I
and Valiant II,
isolated the plate
tuning capacitor
from the chassis. E.F.
Johnson used
insulated washers to isolate
the frame and shaft,
and grounded the
capacitor only through a
long thin lead to a
single point
connection at the PA
tube socket.
Typical insulated
hardware.
Tuning capacitor
fames (rotors) mount
on insulated
washers.
Replace them with
stainless steel.
This capacitor
mounting arrangement
is a source of
frequency
instability and
chirp, and even
causes the VFO to
run on a slightly
different frequency
when the rig is
transmitting then
when the rig is on
“spot”. The
capacitor mounting
also causes TVI!
Harmonic and
fundamental current
path caused by:
1.)
insulating plate
tuning capacitor
from the chassis
2.)
insulating the
capacitor front
bushing from the
shield
3.) imperfect
grounds on front
panel
The imperfect
ground on shaft
entrance of VFO
shield allows a
portion of the plate
tuning capacitor
currents to enter
the VFO housing.
This causes the
VFO to change
frequency in
proportion to plate
tuning capacitor
currents. It also
allows harmonics
from the PA tube(s)
to pass through the
front panel and
cabinet shielding,
where they can
radiate and cause
TVI.
I think I
understand the basic
thought-process
Johnson engineers
had. They just
didn’t consider the
impedance of the
ground paths and the
problems they were
adding. In effect,
they unintentionally
added a path that
partially bypasses
the low-pass tank
circuit for
harmonics and
also increased the path
length for
parasitics. The
negatives of doing
this far outweigh
any unlikely
improvement created
by using a
common single point ground.
My guess is one or
more of the original
engineers grew up
with wooden chassis
breadboard radios
and transmitters.
Old wooden chassis
radios used a single-point RF
ground. This was
because there was no
large highly
conducting
groundplane (the
metal chassis)
available.
The single point RF
ground
practice was very
common, and was
often a requirement
in wooden chassis
equipment! Without a
single ground point
for each stage, old
radios could be very
unstable.
Unfortunately this
practice does not
transition into
metal chassis
equipment,
especially when
shields and extreme
amounts of gain are
involved.
The chirp and
instability (and
TVI) problem is
rooted in the high
circulating current
of the tank, and the
fact the tank is
grounded to the
chassis only by a
single thin wire
connection at one
end of the capacitor frame.
Viking Valiant chirp
and TVI mod
The plate tuning
capacitor shaft
routes through but
is
insulated from
a metal shield.
Shafts should always
be grounded to
shield walls,
otherwise they are a
source of conduction
through the shield.
Not grounding a
shaft to a shield
renders any shield
partially or totally
ineffective.
This is also true
for ingress, any
shaft going into a
shield, like the VFO
box, should be
grounded to an
effective RF zero
point. Unfortunately
the front panel is
not an RF zero
point, because the
shield wall is not
an enclosure, and
the partially RF
floated capacitor
shaft couples RF to
the front panel.
Note the
capacitor shaft
bushing is insulated
from the shield, and
the shaft coupler is
NOT insulated.
This is a huge
mistake because if
the frame of the
capacitor has any RF
potential to the
chassis the shaft
will carry RF
current to the front
panel. This means
the capacitor shaft
will radiate
unwanted RF,
including harmonics,
outside the shield
walls.
The frequency
jumping, frequency
instability,
frequency movement
when adjusting
tuning or loading,
and chirp can most
easily be cured by
removing the fiber
insulating washers
under the chassis
and replacing them
with thickly plated
or stainless steel
washers. Use a
little dab of
dielectric grease
under the washer to
prevent corrosion,
or place a star
washer under the
flat washer to
“bite” into the
chassis. Stainless
hardware will not
corrode as readily
as galvanized steel,
and will generally
retain an electrical
connection much
longer.
I could have
replaced the plate
tuning capacitor
shaft coupling with
an insulated
coupling, grounded
the tuning capacitor
shaft, or corrected
the problem in any
number of ways. I
decided this was the
least invasive way
of solving the
problem.
Johnson Ranger and
Ranger II
There is no
reason at all to run
a 5 inch ground buss
on the plate tuning
capacitor.
Remove the
insulators and the
buss wire going to
the tube socket.
Remove both under
chassis washers on
the tuning
capacitor, and one
by the rear of the
bandswitch that has
loading padder
capacitors.
Use steel washers
to ground them
directly to the
chassis.
If you pass this
information along or
steal it for your
own web use, please
give me credit for
the initial work. It
took me several
hours to find this
problem!
copyright 2004
W8JI
|