The Astro 200
The
CIR Industries Astro 200 is a compact, synthesized 100 watt HF
transceiver from the mid-late 1970s that covers the 80, 40, 20, 15 and
10 meter bands. Intended for both home and mobile use, it is quite
small - 9.75" wide, 12.5" deep and 3" tall (24.8 x 31.8 x 7.6cm) -
including the rear heat sink. Back in 1977 - when this unit was made -
it seems to have cost around $995 for the version without the CW filter - about $5000 in 2025 dollars!
Figure 1:The front panel of the CIR Astro 200. While advanced for
its day, the radio is pretty simple by today's standard.
The lack of a tuning knob seems a bit odd.
Click on the image for a larger version.I
don't know too much about CIR Industries, except that it was around
only for a few years, apparently absorbed by Cubic-Swan in about 1978
where it was rebadged with the name of the new company and - with very minor changes - became the "200A". The history of
Cubic-Swan becomes a bit muddy after the early 1980s and appears to have
fizzled entirely by the mid-late 1990s. Much of the design of the
Astro 200 - and other Cubic-Swan radios - was apparently done by Don
Stoner, W6TNS (who was also the "S" in SGC).
The later version of this radio, the Astro 200A,
sported a 6 pin round microphone connector, black knobs, slightly
different switches, a lighted meter and very slightly modified scales
on the meter itself: I suspect that the electrical differences - some of which are noted below - may have evolved during the production of the original Astro 200.
The radio's history
This
unit was purchased new in 1977, with the extra-cost CW filter option,
and owned by a friend of mine, having first resided in his
International Scout II - and then his Jeep CJ-7 - until about 2020 (when it was removed during vehicle maintenance)
seeing many hours and miles bouncing around rough, 4WD roads. Despite
having banged around for about 40 years in a vehicle, it's in remarkably
good physical shape, the case having only a few minor scratches.
Unfortunately, my friend became a silent key in 2022 and the radio ended
up in my hands.
A "unique" radio
The
advertisements for this radio tout it as being the very first
completely synthesized amateur transceiver: Whether or not it's
actually the "first", I can't be sure, and this can vary depending on
what you mean by "synthesized" - but in this case the local local
oscillators are referenced from a single crystal while the BFOs were
independent - a common practice even into the early 2000s. Being an
early synthesized radio, it does have a few interesting quirks:
- There's no tuning knob.
Tuning is accomplished by a pair of "up/down" momentary toggle
switches. At first, this seems awkward, but one can quickly become
adept to tuning a radio this way. My friend (the one who'd owned this radio)
noted that this tuning method was more convenient when bouncing about
on a bumpy Jeep road than trying to use a conventional knob.
- Operating the "fast" switch moves the frequency up/down by about 20 kHz/second after a brief pause.
- Operating the "slow" switch moves the frequency up/down about 400 Hz/second after a brief pause.
- A brief up or down push-and-release of either switch moves the frequency by 100 Hz.
- 100 Hz tuning steps + Fine Tuning.
The radio tunes in 100 Hz steps, but it has a "Fine" tuning knob that
moves the frequency up/down by a bit more than +/-65 Hz to allow one to
get the frequency as close as you wish. With the tendency for most amateurs these days to set their radios to an integer number of kHz (and occasionally to "0.5", 100 Hz steps are just fine and this control can be left centered most of the time.
- The synthesizers are a bit slow to lock.
As one tunes the radio - particularly in the "fast" mode - the
synthesizers may take a second or so to catch up as it "swoops" in onto
the correct frequency. This also means that after power-up, the radio
is unusable for about 30 seconds, or for up to 15 seconds after changing
bands. As the synthesizers "land" within about a second during normal
tuning with the up/down switches, the radio is on frequency by the time
normal human reaction time has "locked in" to what is on frequency.
- The "WWV" mode. You'll note that the mode switch includes a "WWV" position. This is actually a completely separate, direct-conversion receiver - with no AGC - that is tuned only to 10 MHz. Since it uses the (doubled) 5 MHz reference as its local oscillator, it provides an easy way to check/set the radio precisely on-frequency.
Despite having a digital readout and a synthesizer, it does not have a computer of any sort. "Programming" is done using PROMs (Programmable Read-Only Memory) to look up the synthesizer tuning information and "74LS" type logic as counters for the frequency dividers and tuning - but this also means that when it's first powered up, it
always defaults to the bottom edge of the band to which it is tuned.
This is a bit of an inconvenience - but in the mid 1970's, prior to inexpensive
single-chip microcontrollers with onboard program memory along with affordable development tools there was no
real way around this without adding significantly to complexity and
cost. I'm looking into a simple way for the radio to "remember" the
last-tuned frequency on each band - perhaps the topic of a later
article.
About this radio
Figure 2:The radio's tag - Serial #8, apparently!
Click on the image for a larger version.This radio is apparently a
very early production unit - somewhat different from that depicted in the manual:
- The Microphone connector is a standard 1/4" TRS (headphone)
jack rather than a 6-pin round connector apparently used later in the
production run and in a later revision, the 200A. The additional pins on the 6 pin connector
provide up/down tuning and 11 volts, allowing one to do tuning via
the microphone.
- It was lacking the "ANL Board". This is a very simple circuit circuit (two pairs of back-to-back diodes and an electrolytic capacitor)
that reduces, according to the manual, "excessive popping or AGC
pumping". As this circuit is very simple, it was trivial to add to this
radio and this somewhat reduced the tendency for the receiver to be
momentarily deafened when changing modes or bands.
- Upon inspection of the PA (Power Amplifier)
module I noted that the driver transistors were Motorola, marked with
"604/438 Sample" which further implies an early production radio. The
PA transistors themselves - which are shown as being of type MRF454 in
the service manual - were CD3435 made by CTC.
- There
are a number of doubly-balanced diode-ring mixers used throughout.
Based on the manual and photos of other units, these seem to be
implemented with some sort of module. On this unit these modules are not
used as the corresponding areas on the PC boards are populated with a
pair of trifilar transformers and individual diodes comprising the
mixer.
- The
serial number of this radio is "706008". Based on other photos that
I've seen online, this is apparently serial number 8 - likely having
been made in June of 1977. The date codes on internal components are
consistent with the possible June 1977 assembly date.
Figure 3:The inner synthesizer board - a bunch of counters. LS-TTL
circuitry is used extensively, along with a few diode-type
PROMs for frequency/display lookup and counter set-up.
Click on the image for a larger version.EvaluationAs
I had other projects in the queue, it was only recently that I pulled
this radio off the shelf. Prior to setting it on my workbench, I blew
the dust off it and carefully cleaned the front panel and around
controls, throwing the knobs into an ultrasonic cleaner.
Powering
it up, the unit worked - sort of: I could hear noise, but it seemed a
bit deaf - but the sensitivity changed wildly with a bit of thumping on
the case, an indication of a dirty transmit/receive relay. Even
with a massively strong signal into the antenna connector - which
produced a deafeningly-loud tone in the (external) speaker - I got no S-meter reading. Many years ago, my friend and I used this radio (when it was still in his Jeep) and
noticed this same problem and that it was also mitigated by a
"percussive repair" and/or clicking the PTT several times, indicating
that the Transmit/Receive relay may have problems.
Figure 4:The main RF/AF board, post repair. The layout is a bit
crowded, but pretty clean on a two-sided, glass-epoxy board.
This radio includes the optional 400 Hz CW filter.
Click on the image for a larger version.Popping
the top cover I could see that I had some work to do. While it was
remarkably clean inside for having been in a dusty Jeep for decades, I
could see evidence of a few problems: I saw at least one "blowed-up"
capacitor near the audio amplifier.
Fortunately, the
synthesizer itself seemed to be OK: The tuning controls did their jobs
properly, the tone in the speaker indicating that the radio was landing
on the same frequency as the display. The only "digital" problem seems
to be that one of the segments of each digit on the display was
constantly illuminated, weakly, possibly indicative of a problem with a
segment driver.
Refurbishing
The
first order of business was to replace the electrolytic capacitors. As
a few of them had clearly failed as evidenced by inspection, they all
had to go - particularly since the radio had spent many summers in a
closed vehicle during hot, Utah summers - plus, this radio is nearly a
half-century old (which seems amazing when you consider that it's "digitally synthesized") so time would saved to simply "shotgun" them all. Furthermore, many of the boards are "tethered" with soldered cables: There is just enough slack to pull them out and work on the boards unsoldering only a wire or two, but doing so many, many times would not only be tedious, but risk fatiguing and breaking them - another reason to replace the capacitors in just one session.
Figure 5:VCO/Synthesizer board. There are two synthesizers - one
for them provides the 100Hz tuning steps. Again, LS-TTL
logic is used, along with a few op-amps.
Click on the image for a larger version.Capacitors, and more capacitors! I
took inventory, inspecting the entire radio and come up with the
following list of capacitors - including those found in the PA module
and places other than on the PC boards:
- (5) 470uf, 10 volt
- (2) 330uF, 16 volt (axial)
- (2) 220uF, 16 volt
- (14) 100uF, 16 volt
- (13) 33uF, 16 volt
- (2) 10uF, 25 volt
- (4) 10uF, 16 volt
- (8) 4.7uF, 25 volt
- 1) 4.7uF, 16 volt
- (9) 1uF, 25 volt
- (4) 1uF, 50 volt
- (3) 1uF, 50 volt (axial)
- There
are several dipped tantalum capacitors in low-level voltage and signal filtering lines that seem to be OK for now. As none of these are on power rails there's no chance of a catastrophic
failure (e.g. flames) should one short out. These capacitors will be replaced in the
future.
I suspect that the differing voltage ratings of some of the same-value capacitors was likely to save space (lower-voltage capacitors are generally smaller) and allow the use of less-expensive capacitors, but these days, capacitors are much smaller (and cheaper, in equivalent money)
than their decades-old counterparts. When ordering replacement
capacitors I simply got same value rated for at least the voltage of the
highest in the list above, but the new capacitors also had a
temperature rating of 105C rather than the 85C of the original - and since modern capacitors are smaller than those from about 50 years ago, even a higher-voltage new capacitor was smaller than the lower voltage devices of the same value. Since
the electrolytic capacitors were pretty inexpensive - typically less
than US$0.10/each for the smaller values - I ordered more than just the number above (in some cases, many more) in the event that I missed something.
Figure 6:The pile of electrolytics removed from the radio.
Replacing
every capacitor was the right choice!
Click on the image for a larger version.Removing
capacitors en masse is best done with the appropriate tools -
particularly on an older circuit board. Fortunately, I have a Hakko
FR-300 desoldering iron/pump which made removal much easier and I was
able to avoid damaging any traces on the board.
When replacing a
bunch of capacitors, I prefer to do so methodically, moving from section
to section on the circuit board - noting the polarity orientation of
the capacitor before removing it and if there was any
doubt as to which way it went, referring to the board layout diagram in
the service manual - particularly since the circuit boards have neither
solder mask or silkscreen as a visual reference. Once a capacitor is
replaced, I typically mark the top of the can with a colored permanent marker to
help make sure that I don't miss any.
One possible "gotcha" was that unlike modern electrolytic capacitors which are typically marked only
on the negative lead, many (but not all) of the original capacitors in this radio had
only their positive side marked - which was the custom of some
manufacturers of the day - so I had to be particularly careful to
identify the polarity correctly as I replaced each capacitor.
When
I was done, the receiver seemed to be more "alive" than before, but it
was still a bit deaf - and the synthesizer seemed to be a bit "wobbly",
being very sensitive to slight changes in power supply voltage. The
biggest change was the WWV receiver which was profoundly deaf prior to
the capacitor change-out, but "normal" afterwards.
Capacitor brand implies longevity
After replacing the capacitors I went through the pile and found that most
of them were "OK" - or at good enough that their respective circuits
would have worked. The brand seemed to be a pretty good indicator of
which was likely bad: The Japanese blue-label Nichicon and gray "Sun"
and "Elna" brands were generally OK, the silver and gray Taiwanese
"T.I." brand were all over the map, the "Sam Hwa" and "Towa"
capacitors were marginal, but all of the "Temple" branded capacitors (which seemed to have 1970 date codes - apparently already a few years old when the radio was made) were extremely bad.
After doing this I still believe that replacing all
of the electrolytics was, in fact, the correct choice as I would have
probably spent more time finding and diagnosing capacitors individually -
and possibly suffered near-term failures - than simply swapping them
all out.
A wobbly power supply
With all
of the electrolytic capacitors replaced, I systematically went through
the adjustment steps found in the user and service manual (which can be found online) - more or less. Knowing that before you make ANY
adjustments that you must make sure that the power supply is correct, I
probed about with a volt meter noticing that the 11 volt supply was
actually just below eight volts, likely accounting for its seeming
deafness. Locating the 11 volt regulator on the synthesizer board, I
noted that the act of slightly adjusting the potentiometer resulted the
voltage jumping, indicating that it was somewhat "stratchy", with the
wiper likely not making good contact. A bit of cleaning spray and
exercising of this control resolved the issue and I reset the voltage to
precisely 11.0 volts.
Figure 7:Original S-meter coil. It would seem that the coil winding
was broken in several places - hence, unsalvagable.
Click on the image for a larger version.With the correct voltages now applied to the circuits in the radio, its sensitivity seemed to be
much
better and the synthesizer was no longer sensitive to fluctuations in
the power supply, being able to tolerate a drop to about 11.25 volts at
the radio's DC input before the synthesizer "wobbled".
No S-meter! Going
through the alignment steps I applied a signal from my generator and
noted that while the sensitivity seemed to be about right - and the AGC
was now working as it should - the S-meter did not move. It's worth
noting that the S-meter on this radio works ONLY when the
meter switch is set to the "ALC" position - but I was getting no reading
on any setting. Using a voltmeter, I could see that the voltage across
the S-meter's movement was increasing with the signal strength
indicating that the AGC was working (which was also obvious by listening to off-air signals) but a quick check with an ohmmeter - after disconnecting one of the meter's leads - indicated that it was open circuit.
This
was bad news, particularly since it was likely that I would never find a
meter of the same, exact physical size - and even if I did
find a replacement, I'd probably have to re-create the scale in the
meter. This wasn't impossible to do, but I took another path.
Figure 8:The meter with its rewound meter coil using #30 wire.
As many turns were wound as would fit - the coil shaped to
prevent mechanical interference and then covered with
varnish to hold it in place.
Click on the image for a larger version.Carefully
disassembling the meter and inspecting it I noted that it was of the
inexpensive "moving vane" type, the coil wound with very fine wire - probably around 46 AWG. In
probing very carefully I noted that one of these hair-thin wires was
disconnected at the base of the coil. Further probing showed that the
wire itself was frayed where it was wound onto the phenolic paper stator
- probably a victim of both temperature cycling and (possibly)
some corrosion. A bit of later inspection of the wire showed that it
seemed "brittle" - something that I've seen on older gear: I don't know
if it's the copper hardening in some way or some sort of reaction
between the wire, enamel and its environment that causes this.
Since
the meter's coil was a total loss I decided to do something a bit
drastic: Rewind it. Rather than trying to use #46 wire, I chose,
instead, to use less-fragile wire - #30, which is about 10 times larger
diameter: I'd have used a smaller - but not overly fragile - wire (likely #36) if
I'd had it on hand to get more turns and better sensitivity. Of course,
I was not going to get nearly as many turns on the stator as the
original - which meant that it wasn't going to be as sensitive as it had
been originally and would be unlikely to work properly in the circuit -
but I had a plan for this.
Carefully winding the #30 wire into
the phenolic stator until it was "full", I scrunched the coil down to
reduce its height and then pushed it sideways to clear both the meter's
axle and the moving magnets on the rotor before covering all of the
windings with urethane varnish. With the varnish dry, I reassembled the
meter and using a variable bench supply with a series resistor to vary the current through it I found that it operated nonlinearly, particularly near the
upper and lower ends of meter travel.
I quickly realized that the
screwdriver that I'd used was slightly magnetic - and the two screws
used to hold down the phenolic stator had become magnetized as well from
using that screwdriver. Using a TV picture tube degaussing coil (I could have used a soldering gun's magnetic field instead) I demagnetized the two screws and the screwdriver, solving this nonlinearity problem.
Re-zeroing
the meter and using a series 470 ohm resistor and a variable bench
power supply I found that the meter's full-scale sensitivity was about
23 milliamps - very much higher than the 500-ish microamp sensitivity
that I'd calculated it to be originally. In looking at the circuitry I
noted that the negative side of the meter was grounded in all three of
the front panel meter switch settings which meant that all I needed was
to come up with a circuit to multiply the current linearly - and with one end of the meter being connected to circuit ground, that task was greatly simplified: Here's the circuit to do this:
Figure 9:Schematic of the circuit used to drive the re-wound meter on the CIR Astro 200.
The
circuit shown in Figure 9 is the classic "precision current source"
using an op amp to drive a transistor and then the meter. The input
voltage is scaled with the trimmer potentiometer (R3) and applied
to the non-inverting (+) input of the op amp with R4 in parallel to set
an input resistance of about 250 ohms - which is my guess of the
resistance of the original meter movement. By its nature, the op amp
will attempt to adjust its output to make the voltage on the inverting
(-) input the same as the non-inverting (+) input and to do this, it
turns on the transistor, causing current to flow through the meter and
the current sense resistor, R2. Resistor R1 is there to limit the
maximum current to a "sane" value to prevent the meter from being
slammed too hard in the case of an "oops".
Figure 10:The as-built circuit from Figure 9 constructed on some
prototyping board. This circuit is adhered to the top of the
meter itself.
Click on the image for a larger version.The
result of this is that this circuit will happily convert the voltage
through R2 into a proportional current, the magnitude set by the
adjustment of R3, allowing our now-rebuilt meter movement of arbitrary
sensitivity to be used.
As the schematic shows, this
circuit was built using the venerable LM324. This device was chosen
mainly because I have plenty of them, and it's one of the most common op
amps that has an input and output voltage range that includes "ground" (V-):
Many "standard" op amps don't work near one or the other power supply
rail and will work incorrectly if the input voltage is the same as the
"V-" lead (ground, in our case) and about as many cannot output voltage down to the negative rail, either.
Since
I needed only one of the four LM324's op amps, the other three were
simply strapped to the power supply to keep them from floating and
possibly causing noise issues: It's possible that I could have used one
or more of the op amp sections to directly drive the meter, but the
single transistor was cheap and easy. The circuit was built onto a
small piece of glass-epoxy perfboard and attached to the top of the
meter movement - the power supply from this circuit stolen from a trace
containing the +11 volt supply found on the front-panel circuit board -
but even the 13 volt, unregulated supply would have been fine.
Setting up the "new" meter
While the actual sensitivity of the original meter - which is believed to be around 500 microamps - is not known for certain, there is one
step in the manual that is revealing in that it has no actual circuit
adjustment, relying on the sensitivity of the meter itself and fixed components for
accuracy and calibration. Because of this, we must do this step first and calibrate the sensitivity of our new meter circuit.
In the section of the manual about "Power Meter, Reflected Power Meter Adjustment" it describes connecting a 2:1 VSWR load (25 ohms using two 50 ohm dummy loads in parallel)
and using an external power meter connected between the radio and the
load: The radio should be set for 40 meters for this step. Switching
to "CWW" (CW Wide - using the SSB filter) mode, set the Mic Gain to maximum (fully clockwise), key the radio and then increase the power (turning the Mic gain counter-clockwise to increase power) and adjusting R312 to limit the maximum power to 90 watts even when the Mic gain control is fully counter-clockwise (maximum power):
These adjustments should be done quickly to avoid overheating the power
amplifier. The manual notes that with the meter set to the "REF"
position, the meter should read "2" (for 2:1 VSWR) - and we quickly adjust R3 in Figure 9 for a reading of "2" on the meter.
Again, the key point here is that the REF meter gets its output from
the reverse power detector amplifier - but since its threshold is fixed,
when the power is being reduced by this circuit, it will always
output the correct voltage/current to make the meter read "2". In other words,
this is fixed reference and we can use it to calibrate the meter for
all other modes.
After this, the procedures for adjusting the S-meter, ALC and forward power readings outlined in the manual should be applied without further adjustment of R3, the 10 turn trimmer potentiometer on our meter-driver circuit from Figure 9.
It's
worth reiterating the point that as the AGC, ALC, FWD and REF signals
feeding the meter are ground-referenced, the circuit design was simple.
If the
meter was driven by a "floating" circuit - one in which the negative
side of the meter was at some potential other than ground - I would
likely have used several sections of the LM324 configured as an
"instrumentation amplifier" - one that measured the voltage drop across a fixed resistor (in lieu of current through the original meter coil) regardless of the actual voltages. This circuit would have been somewhat more complex, but not overly so.
Radio alignment
With
the capacitors replaced and the meter working, I went through the
alignment steps outlined in the manual. Fortunately, I had reviewed the
manual in its entirety and noted a few "inconsistencies", notably:
- The
listing of the carrier oscillator frequencies in the alignment steps
shows the same frequency for LSB and USB. The correct frequencies are
shown on the previous page.
- When adjusting the ALC using potentiometer R296, the manual says to do so at mid-rotation in one place and and fully CW (clockwise) in another: I presume that they meant fully CW.
Additionally, I would suggest the following additions to the procedure at the beginning of the procedure.
- Verify/adjust the setting of the 11.0 volt regulator on the synthesizer board (R92).
- Verify/adjust the 5.0 MHz oscillator on the synthesizer board using C52.
- If you had to re-wind the meter and add the circuit described above, I would do the reverse power meter calibration (described above) before the other meter calibration steps: This is noted in the procedure at the end of this article.
After this, proceed with the alignment/calibration as described in the manual. There is a revised/annotated alignment procedure at the end of this article.
Power cable
As I was unable to find the original power cable (it may still be in the Jeep)
I needed to find the mating power connector. Recognizing it as a
"Jones" connector, I did a bit of research and found that I needed
to get a Cinch-Jones S-306-CCT, which is a 6 pin female connector.
Unfortunately, this line of connectors was discontinued by the
manufacturer several years ago, but EvilBay came to the rescue and I
found a "new" one with the inline cable shroud and strain relief.
Using
12 AWG wire and an inline holder with a 30 amp blade fuse I put
together a power cable with an Anderson power pole connector on the far
end. This allowed me to connect it to a high-current power supply so
that I could get on with testing the radio's final power amplifier.
"Final" problems
With
the radio otherwise aligned, I noted that I was unable to get anywhere
near full power out of the power amplifier - about 35 watts on 80
meters, nearly 50 watts on 40 meters and 10-15 watts on 10 meters.
Checking the output on the main RF/AF board, I noted that the voltages
were equal to or higher than noted in the manual so I removed the PA
module from the back via its ten screws.
I immediately noticed
something that further indicated that this was an "early" unit: The PA
driver transistors were Motorola, but marked as "604/438 Sample" and
rather than using MRF454 outputs, they were CTC CD3435. In poking
around with an oscilloscope with about 10 watts of output on 40 meters I
noticed that the waveforms on the collectors of the driver transistors
were not equal - and neither were the corresponding waveforms on the
output transistors: This indicated that in each stage, at least one of
the transistors had failed - or was badly degraded.
While annoying (the transistors aren't cheap!) it didn't surprise me. It is (apparently)
common for RF transistors from the 70s and, perhaps, into the early 80s
to fail - even when not being used - due to internal defects that seem
to "grow" over time.
Figure 11:The repaired PA board with the new driver and output
transistors.
Click on the image for a larger version.For
the driver transistors, the originals were 2N6367, but the equivalent
is the MRF433 or the 2SC2395 - but the MRF455 may work OK. Rummaging
around my bin of RF transistors I found a pair of pulled 2SC2395s
(I don't recall where I got them)
and put them in, saving me from spending about $100 for them. Greeted
with only about 80 watts on 40 meters - and much lower power than that on 10 meters - I could still see from the waveform on
the 'scope - probing the collector leads - that one of the output
transistors was still an issue.
While I could get a pair of
MRF454 transistors from RF Parts, I noted that they were available from
Mouser Electronics for a lower price (about $55 each at the time of writing) and when
they arrived, I saw that they sported a recent date code. Plopping them
in I saw that the PA was now capable of well over 125 watts on 80 and
40 meters - working as it should - allowing me to complete the
adjustment procedures related to the ALC and power metering.
In
testing the two original PA transistors out of circuit I noted that
both their beta and "diode drop" voltage were radically different. I
suspect that at least one of these devices had lost some "emitter sites"
or tiny bond wires on the die, making it "less of a transistor" than it
once had been.
With the final board now repaired, the radio met the specifications
outlined in the manual: 100+ watts on all bands except 10 meters where
the output was a bit over 85 watts.
A few loose ends...
The "stuck" LED segment
I
also noted that the "stuck" segment on the LED display seemed to have
fixed itself during a toggle of the "bright/dim" switch: In looking at a
YouTube video reviewing this radio I noted that it, too, had this exact
problem with the same segment being stuck - but I have no idea if it's common (e.g. happened on at least two different radios) or why it fixed itself - nor is there an obvious clue from the schematic diagram why that one particular segment would be affected on my radio and the one in the video.
Adding the clipper/limiter
As
for the receiver, the sensitivity is good - but I decided to make a
modification that apparently became standard in production just after
this unit was produced. I noted that when changing modes and bands, the
S-meter would "pin" with the very loud "pop" that occurred, the AGC
taking 5-10 seconds to recover
Figure 12:The clipper circuit in tubing, installed in the radio. One end
is connected to a leg of R290 - the other end to ground.
Click on the image for a larger version.Noting
that the manual included the description of a "Limiter" board - and
that the radio in the YouTube channel - which had a serial number of
about a dozen units higher - also had this board, I figured that this
might be one of the reasons why it was added.
The circuit itself is simple: Two pairs of diodes - one silicon and one germanium in series (for a clipping voltage of about 0.9 volts)
- were placed in anti-parallel configuration and AC-coupled with a 10uF
capacitor. This circuit was placed between ground and input of the AGC
detector. Rather than make a small circuit board as was done in the
production units I simply wired the components in free space and covered
them with PTFE and heat-shrink tubing, connecting the assembly between
the AGC circuit and a handy ground pin as can be seen in Figure 12.
My
suspicion about its later addition was confirmed: While there is still
a loud "click" when changing modes, the AGC now recovers much more
quickly and the radio's AGC is also very much less prone to being badly
deflected with a long recovery time when there is a loud static crash.
The T/R relay and filter module
Mentioned
briefly, there was the problem with the intermittent T/R relay. This
is contained within a module that sits along the right edge, inside the
radio that extends from the front panel to the back of the radio along
with the band switch.
This module - in addition to the T/R relay -
contains the receiver pre-selector filters, the transmit mixer filters
and the transmit low-pass filters on a compact, shielded assembly. To
pull this assembly out of the radio would be quite a job, requiring the
partial removal of the front panel, disconnecting (mostly unsoldering!) a number of wires, connectors and signal cables and pulling it out of the radio - something that I have not attempted to do.
Fortunately, the designers provided an access hole near the back panel of the radio (on the bottom side)
that is covered with tape where one can burnish the relay's contacts and apply contact cleaner.
After both burnishing and the application of cleaner, the T/R relay is
now working perfectly.
Using the radio
Tuning with switches
With
the use of toggle switches instead of a round, "spinny" tuning knob,
operating the Astro 200 is decidedly different than using a conventional
radio. As mentioned before, the previous owner told me that he thought
using toggle switches was a bit better for tuning while bouncing along
bumpy roads than a large knob - and in the days of analog radios, this
was likely the case.
In perusing online references to this same
radio, the users also noted that one quickly becomes accustomed to this
method of tuning - but everyone had the same comment: It's slow to tune
across the band. When powered up, this radio always starts at the
bottom of the selected amateur band - and on 10 meters, this particular
radio starts at 27.0000 MHz (transmit is inhibited below 28 MHz) which means that it takes about a minute to even get into the 10 meter band!
The AGC
The
radio's AGC is not adjustable and the time constant is fine for CW, but
a bit fast for SSB in my opinion. As is common with many analog
radios, the apparent AGC time constant gets shorter with more AGC action (e.g. higher S-meter reading). This is a result of the "dB per Volts" curve getting steeper with many gain reduction schemes (e.g. more dB gain reduction per volt of change) effectively shortening the time constants.
Since
this radio has a front panel RF attenuator control, switching this in
to reduce the signal level helps with this effect somewhat.
Noise blanker
The noise blanker (enabled by pulling the "Squelch" control knob out)
seems to work pretty well, operating in the wideband IF prior to the
crystal filters. As is typical with noise blankers in analog receivers - and some
modern digital radios - its efficacy is somewhat affected by very
strong, adjacent signals which "desense" the noise detector - a
difficult problem to overcome.
CW usage
As is
common for radios of that era, the sidetone frequency in the CW mode has
little to do with the frequency offset. This radio uses USB and a
positive transmit frequency shift when in CW which means that neither
the display or the tuned frequency changes when going from USB to CW
mode. This was pretty common in the era (many makers - including Drake - did it this way)
which meant that if the operator wanted to know the actual frequency of
their transmitted signal that they would have to do some mental math.
One
"quirk" that I need to investigate is that if this radio's heterodyne
oscillator is set precisely according to the manual, the receive (and transmit)
frequencies do not match the display, being offset by a bit more than
100 Hz. This is easily corrected by setting the display to a known
frequency, inputting a signal 1 kHz above and below (for USB and LSB, respectively)
and adjusting for an audio tone of 1 kHz, but doing so shifts the
passband of the crystal filters audibly - and in CW mode, it puts the
center of the passband at about 1200 Hz. This slight shift does not
result in either "tinny" or "muffled" audio when using SSB on either
sideband, and the radio sounds quite good on air!
As this
offset - which is mentioned in the manual as being around 1000 Hz -
appears to be programmed into PROMs, it does not seem possible to shift
the local oscillator to overcome this issue - and while there's a
difference between the USB and LSB passband, it is not a "show stopper"
but a 1200 Hz-centered passband for CW is too high in my opinion. I
suspect that this being a very early production radio may have something
to do with this issue and I'll have to think about possible ways to
address it.
When a "Mic Gain" control isn't really "Mic Gain"
Another
unusual design feature of this radio is the transmit audio path. From
the microphone input the signal path goes directly to the amplifier (there's no level adjustment preceding it)
and into the clipper/compressor stage. Interestingly, the
clipper/compressor takes the form of a logarithmic amplifier which has
less of a sharp "knee" than a typical clipper, making it quite effective
in functioning very much like a compressor-type speech processor.
The designers made an interesting choice here: The control marked "Mic Gain" is placed in the signal path after
the clipper/compressor - but this has some important implications. In
testing, I used an old Sure 440SL high impedance dynamic microphone
which has a fairly high output level, but this caused the
clipper/compressor to be "hit" very hard: On-air reports indicated that
that I was readable, but that my speech processing was very "heavy" and
off-air recordings from a remote WebSDR verified this. Since the "Mic
Gain" control is between the clipper/compressor and the radio's balanced
modulator, it affects only the RF output power and how hard one is
"hitting" the ALC and doesn't affect the amount of audio compression at
all.
What should really have been done was to include a
means of adjusting the microphone level into the clipper/compressor
stage and this could take the form of having a level control on the
microphone itself or in a box between the microphone and the radio, or, if the same microphone will always be used with the radio, put such a control inside the radio.
To
accommodate this need, I rummaged around my parts box and found a 500k
vertical chassis-mount trimmer potentiometer. This potentiometer was wired
such that the "CCW" (counter-clockwise) end was grounded and the opposite end connected to the microphone jack with the audio to the radio on the wiper.
Figure 13:A 500k potentiometer - reachable using a long, thin blade
screwdriver is accessible through the 1/4" TRS MIC/KEY
connector on this radio.
See text for more details.Click on the image.As
depicted in Figure 13, behind the 1/4" MIC/KEY jack is a 5 volt
regulator in a TO-3 case - but this doesn't line up with the connector,
so I glued the potentiometer to a small piece of circuit board to allow
it to be offset. When I glued the pot to this board, I took care to
avoid fouling the adjustment knob and after curing.
I then glued
the small piece of circuit board to the top of the 5 volt regulator,
taking care to offset it so that the potentiometer was aligned such that
a long, thin blade screwdriver through the MIC connector could be used
to adjust the level from the microphone being applied to the MIC
amplifier.
The adhesive that I used was "Shoe Goo" which remains
flexible: I would not recommend epoxy, cyanoacrylate ("super") glue or hot-melt glue as none of these are a good choice in this application (e.g. the bonds will fail with temperature cycling and/or mechanical stress.)
Adjusting
this new "MIC Level" control is an iterative process: Plug in the mic -
check the ALC deflection and output power, unplugging, and then making
the necessary adjustments and doing it again. The goal here is to have enough audio to
activate the compressor, but not so much that it sounds very "heavy"
on-air.
As noted earlier, THIS radio uses a 1/4"
TRS connector rather than the round, multi-pin connector used on later
production models: If this radio had this latter connector, blocking
access to an adjustment behind it, I would have mounted the
potentiometer facing down and drilled an access hole in the bottom of
the chassis, probably making a right-angle bracket on which it could be
mounted.
Carrier balance
One interesting omission
by the designers is the lack of a "carrier balance" control. When SSB
is generated, the "balanced modulator" - which literally mixes the audio
with RF - this carrier is nulled on most radios via one
or two adjustments to minimize the amplitude of the original carrier -
but not on this radio. This radio uses a diode-ring type of
doubly-balanced mixer and by themselves these typically have a
"bleedthrough" of between 25 and 35 dB- much less than the 40-50dB of a
typical balanced modulator in an analog SSB transmitter after it has
been carefully nulled.
What this means is that on 40 meters there is a carrier bleedthrough of about 200 milliwatts (which varies depending on whether you are using USB or LSB, with band and operating temperature)
when keyed down with no transmit audio. Compared with a 100 watt
output level, this represents a level that is 25-30 dB below peak power
that cannot be adjusted. This is nowhere near enough to impair
efficiency of the transmitter by "wasting" power in the carrier but it is enough to be easily visible to the "waterfall police" using a modern digital radio with a spectral display if the conditions are good.
Frequency (in)stability
When
it came out, this radio was remarkable compared to its contemporaries
in that it didn't really drift: You set the frequency and it just stayed there, within a few Hz. Unlike most radios of the
day, it moved only a few Hertz from the instant that it was turned on
while most others at the time would change by hundreds of Hz in the first half hour or so - particularly if operated in a cold environment.
Compared to today's radios, the synthesizer is a bit crude - it has large (100 Hz) tuning steps and a bit slow to lock. As the radio uses rather low reference frequencies (100 and 163 Hz)
for its two synthesizers, their oscillators are rather slow to respond -
but this also means that they are easily disturbed by slight changes in
power supply voltage, mechanical vibration and just the physics of
electronic circuits.
What this means is that the frequency can
easily "wobble" a few Hz - or even 10s of Hz - around the nominal frequency in the short term. This is
generally unnoticeable for SSB usage or even RTTY - and most people will
likely not even notice this when running CW - but it does make this radio unsuitable for some
of the very narrow digital modes that are seen today, like FT-4, FT-8,
WSPR, PSK31 or similar - a trait that it shares with some of its non-synthesized (VFO-only)
predecessors. These modern digital modes require that the radio be
stable within 1-2 Hz at any given instant over the duration of the transmission/reception
window - and this radio simply may not be able to do that.
Mechanical work
If
you look very closely at Figure 1, you'll see aluminum brackets on
either side of the front panel that were used to screw it to the
underside of the dash on the CJ-7 in which it was mounted. During my
refurbishment, I drilled out the pop rivets on these brackets and filled
the holes - and a few scratches - with metal-filled epoxy and sanded
them down.
Even though the exterior of the case was in reasonable
shape, it did show a bit of the wear of having been in a vehicle or two
for over 40 years, so I decided to repaint it. Having been in the
vehicle for so long, the original light blue color was varied, depending
on how much sun had faded it, but inside the top cover - out of sight -
was a "virgin" section of paint to which I was able to find a very
close match at the store: Rustoleum satin "French Blue". Just in case I
- or someone else - wanted to match the original color, exactly, I
masked off and left a patch of the original paint inside the lid.
Aside
from a bit of wear on the knobs and slight yellowing of the panel
meter's clear plastic - most of which was removed with the application
of a bit of elbow grease and Novus plastic polish - the radio looks
almost brand new.
On the air
I've
made several contacts on the air with this radio and and have gotten
good reports. Even with the prevalence of waterfall displays these
days, few people mention the slight carrier leakage - but I also wonder
how many people actually look at their waterfall not to mention how many
others would immediately recognize carrier leakage, anyway?
The
addition of the "MIC Level" potentiometer was a good one. When properly
adjusted, the radio now sounds "normal" rather than very heavily
"compressed" as before.
I haven't used the radio enough to
become very adept at quickly tuning across the band using the UP/DOWN
toggle switches, constantly overshooting signals - but I'd guess that
this would be a skill that could be readily acquired. At the risk of
sacrilege, I'm considering the addition of a small, PIC-based
microcontroller board that will track the button presses and the current
band selection to "pre-set" the frequency when the unit is powered up
and band is changed, making it a bit more convenient to use: Such a
modification would be completely reversible
Final comments
One
should treat this radio in a way similar to "vintage" radios of decades
gone by. It's remarkable in its capability and design considering that
it's nearly a half-century old and it needed relatively little repair - and even more remarkable in that most of the parts that it uses are still available from electronics suppliers at the end of the first quarter of the 21st century - not something that is likely to be true of today's radios in 50 years!.
As
a general-purpose radio for SSB, CW and even RTTY operation, it's still
very usable: It's small size belies its capabilities, particularly in
context with its vintage. Being made prior to 1980, it obviously lacks
the WARC bands (30, 17 and 12 meters) - but so do other radios of
that time period. Once the radio was restored - mostly a matter of
replacing electrolytic capacitors - it operates pretty much as it did
when it was new and it would not seem out of place on the air among
modern radios on the air.
Given its quirks (no tuning knob being the most obvious) it is a bit of curiosity, reminding the user of a time just before completely
analog radios gave way to synthesized radios becoming the norm - a
revolution not too dissimilar to the more recent trend of "analog"
radios giving way to those that are almost entirely digital from the
antenna port to the speaker.
* * * * * * *
Alignment notes
Here are notes related to aligning the Astro 200 (Non "A" version) - although they should be generally correct for the "A" version as well. These should be used to augment the instructions noted in the operation/maintenance manual.
Power supply check - IMPORTANT!
- Verify 11.0 volt power supply - adjust R92 on synthesizer
board as appropriate.
- Verify that 11.0 volt supply will remain stable down to a
supply voltage of at least 11.5 volts as measured on the radio's
voltage input.
- Verify 8.0 and 5.0 volt supplies (each being +/- 0.25 volts of nominal). Note that there are two separate 5.0 volt regulated supplies.
Reference (Master) oscillator:
- Frequency counter to set to 5.000000 MHz or use WWV setting
(which listens to 10 MHz via a direct-conversion receiver) and listen for zero beat
- Set C52 for 5 MHz, exactly. This is accessible via a small hole in the bottom cover.
Carrier oscillator:
- USB/LSB
- MIC Gain CCW
- RIT and FINE at 12:00 position
- MODE to USB
- Key radio and adjust C180 for 5.601650 MHz
- MODE to LSB
- Key radio and adjust C174 for 5.598350 MHz
- MODE to CWN
- MIC Gain fully CW (for minimum CW TX power) and connect radio to dummy load.
- Key radio and adjust C204 for 5.60060 MHz
RX Delay adjustment - used to delay time between release of
PTT/VOX and RX activation
- Adjust R239 for desired delay time preference in switching from TX back to RX when PTT is released.
VOX Trip and Anti-Trip
- Turn on VOX and set volume to desired level using your typical ham shack speaker/audio environment.
- Adjust R181 for VOX activation level with normal speaking
voice.
- Adjust R158 for anti-VOX level with signals/static present to prevent unwanted triggering.
Meter adjustments. Be sure to view meter "straight on" and
consistently to minimize parallax for the readings below.
- VSWR shutdown/reflected power: R312 calibrates the VSWR
shutdown of power. DO THIS STEP AS QUICKLY AS POSSIBLE. Be sure to view the meter "straight on" to avoid parallax in the following steps.
- NOTE: As mentioned earlier in this article, I had to "repair"
the meter by re-winding its coil and using an external driver circuit. If you restore the meter in this manner, do THIS step before the other "Meter adjustment" steps.
- Connect two 50 ohm loads in parallel for 2:1 VSWR (25 ohms)
- use the shortest length coaxial cable possible.
- Set to a mid-band frequency on 20 meters.
- Set meter switch to REF
- In CWW mode, turn MIC gain fully CW, key transmitter.
- Increase power. Quickly adjust R312 so that the forward power can
not be increased to more than 90 watts on the forward meter and unkey.
- In VSWR mode, the meter should read about 2.
- Forward power: R306 calibrates forward power reading.
- Connect 50 ohm dummy load and power meter.
- Set the radio to a mid-band 40 meter frequency and pre-set
the MIC gain control fully CW to set minimum power.
- In CWW mode, turn MIC gain CCW, key transmitter and adjust
for 100 watts on the power meter.
- Adjust R306 for full-scale indication indication (to the
"Set" marking) on meter.
- ALC Setting. Be sure to view the meter "straight on" to avoid parallax in the following steps.
- Connect 50 ohm dummy load and power meter.
- Set MIC gain to 12:00 position, meter mode to FWD. (CONFLICT:
Manual says says fully CW in earlier section about
adjustment)
- Note: Since the transceiver has
no actual "Microphone Gain" adjustment prior to the clipper, the
fully-CW adjustment setting would make sense as it will maximally drive
the ALC (worst-case).
- Key transmitter and whistle or produce tone into the
microphone.
- Adjust R296 for a reading of an average of 40 watts on the
power meter. This should correspond roughly with a reading of
"30 over" on the meter.
- ALC Meter setting. Be sure to view the meter "straight on" to avoid parallax in the following steps.
- Connect to 50 ohm dummy load.
- Set mode to CWW, meter to ALC and set MIC Gain fully
CLOCKWISE
- Key transmitter: There should be low/no power.
- Adjust
R291 for FULL SCALE ALC meter deflection.
AGC set-up. Be sure to view the meter "straight on" to avoid parallax in the following steps.
- Connect signal generator to antenna input and mode to CWW.
- Set front attenuator switch to OFF (down)
- Set for 20 meters and tune to a frequency mid-band and adjust
the signal generator so that there is a tone of about 1 kHz
- Set the signal generator for an output of 1.5 microvolts
(-103.4dBm)
- Adjust R280 for an S-meter reading of S3.
- Increase the signal to 50 microvolts (-73dBm)
- Adjust R272 for an S-9 meter reading
- Re-check the steps above for 1.5 and 50 microvolts and adjust
as necessary.
Sidetone Level set
- Connect to 50 ohm dummy load, set to CWN and adjust MIC gain
control fully CLOCKWISE (minimum power)
- Key transmitter and adjust R257 for desired sidetone level in
speaker.
In-depth alignment:
Carrier oscillator peaking
- Using and oscilloscope or high-impedance RF voltmeter, measure
the amplitude at the base of Q60
- Adjust L11 for maximum amplitude. Use only a plastic adjustment tool to avoid breaking the core.
- Check carrier oscillator frequencies as noted above - adjust
as appropriate.
TX mixer and ALC attenuator
- Connect 50 ohm dummy load.
- Set to CWN and adjust fully CCW (max power)
- Key down and adjust L6 for maximum signal on collector of Q20 using an oscilloscope or RF voltmeter. Use only a plastic adjustment tool to avoid breaking the core.
WWV receiver adjustment
- Set MODE switch to WWV and turn AF gain all of the way down.
- Apply signal generator at 50uV (e.g. -73dBm - equivalent to S9) to antenna, offset from 10
MHz by about 1 kHz so that a tone will be heard.
- Connect AC voltmeter to speaker and adjust level to indicate
on meter, but keep it well below clipping.
- Tune L15 for maximum speaker output. Use only a plastic adjustment tool to avoid breaking the core.
- Remove input signal.
- Using a high-impedance RF voltmeter or oscilloscope, adjust
L16 for maximum 10 MHz at collector of Q77. Use only a plastic adjustment tool to avoid breaking the core.
- A signal of 5uV (-93dBm) should be audible.
Noise blanker adjustment
- Connect a signal generator to the antenna input.
- Adjust receiver and signal generator for a mid-band 20 meter
frequency and adjust for a level of 100uV (-67dBm) and an
approx. 1 kHz tone in the speaker.
- Adjust L9 and L10 for maximum voltage on D30. Use only a plastic adjustment tool to avoid breaking the core.breaking the core.
SWR Bridge adjustment
- Connect two 50 ohm loads in parallel for 2:1 VSWR (25 ohms) -
use the shortest length coaxial cable possible.
- Set MODE switch to CWW and set MIC Gain control fully CW
(minimum power) and set to mid-band on 20 meters.
- NOTE: Do the following measurements as quickly as possible to
minimize stress on power amplifier.
- Key down. Increase power (MIC gain turned CCW) and note that
SWR protection limits to 90 watts as adjusted in SWR protection
steps noted above.
- Note power reading on front panel meter and external wattmeter
(if used) and then un-key.
- In the same manner, check the maximum power into the same 2:1 VSWR
on 80, 40 and 15 meters.
- Adjust C3 as necessary for flattest (most consistent) power
reduction on all bands: Power should be between 80 and 105
watts.
- On 10 meters, power into a 2:1 VSWR may be in the 70-80 watt range.
RF Tuning assembly
This is the unit inline with the BAND switch. The coils noted
below correspond with the frequency range and should be adjust for
best response across that noted below.
NOTE:
As the receive and transmit filter inductors are
not normally accessible, it is necessary to remove the band switch
module to
perform these adjustments - a laborious task which requires
unsoldering a lot of different cables and removal of the front panel.
It should be done
ONLY if problems are suspected. These adjustments should only be
done with a spectrum analyzer and tracking generator OR
a VNA/SNA.
If the sensitivity of the receiver is adequate and the
transmit drive is within specifications, there is probably little need
to even touch these adjustments. As my radio was "up to spec" in terms
of sensitivity and TX drive, I did not pull the module and make any
adjustments.
Use only a plastic adjustment tool to avoid breaking the cores.
Receive filters
- 80 Meters: L101, L102 - 3.5-4.5 MHz
- 40 Meters: L103, L104 - 7.0-7.5 MHz
- 20 Meters: L105, L106 - 14.0-14.5 MHz
- 15 Meters: L107, L108 - 21.0-21.5 MHz
- 10 Meters: L109, L110 - 28.0-30 MHz
- WWV: L111, peaked at 10.0 MHz.
Transmit mixer band-pass filters
- 80 Meters: L201, L202 - 3.5-4.5 MHz
- 40 Meters: L203, L204 - 7.0-7.5 MHz
- 20 Meters: L205, L206 - 14.0-14.5 MHz
- 15 Meters: L207, L208 - 21.0-21.5 MHz
- 10 Meters: L209, L210 - 28.0-30.0 MHz
Synthesizer adjustments
Unless the synthesizer has difficulty locking - particularly at
the upper or lower edge of one or more bands - there's probably no need
to make these adjustments.
Major Loop VCO
- Adjustments should be made at low edge of the respective band.
- Coil should be set for a voltage of 2.5 +/- 0.25 volts on R18
- Exception: For units that can tune to 27.0 MHz, the voltage
should be 3.0 +/- 0.25 volts when tuned to 28.0 MHz.
- Start with the highest band first.
For the progressively-lower bands, the following inductors are in
series meaning that a higher-band coil's adjustment will affect all
lower bands.
- 10M: L9
- 15M: L8
- 20M: L7
- 40M: L6
- 80M: L12
- Notch filter: Adjust R11 and R15 for minimum amplitude of 100
Hz signal on the output (pin 6) of IC21 (0.035Vpp or lower)
Minor Loop VCO
- Adjustments should be made on low edge of the respective
band. (Manual isn't clear about this)
- If adjustment is needed, it will be necessary to remove the
brass shield by unsoldering its three corners. Note that the presence of the shield may affect tuning, so it may be necessary to iteratively replace it during the process.
- Coils should be set for a voltage of 1.6 +/- 0.2 volts as
measured on R21.
- Start with the highest band first.
For the progressively-lower bands, the following inductors are in
series meaning that a higher-band coil's adjustment will affect all
lower bands.
- 10M: L5
- 15M: L4
- 20M: L3
- 40M: L2
- 80M: L1
- Notch filter: Adjust R25 and R27 for minimum 165 Hz signal (0.025Vpp or lower)
on the output (pin 6) of IC20.
* * * * * * *
This page stolen from ka7oei.blogspot.com
[END]
Figure 1:
Figure 3:
Figure 4:
