Radio Boulevard WWII Communication Equipment - Part 3 WWII Ally Radio Communications
Equipment
Kingsley Radio Co. ~ AR7 ~ Royal Australian Air Force "HRO Knock-off" Receiver Marconi's Wireless Telegraph Co. ~ C.R. 300/1 ~ Royal Navy Receiver Canadian Marconi Co. ~ CSR-5 ~ Royal Canadian Navy Receiver
|
WWII Ally Radio Communications Equipment
|
|||||||||||||||||||||||||||||||||
|
Australian Army/Navy
Version - Nearly all of the AR7s receivers that were used by the Australian Army were mounted in a table-style rack
that included a rack mounted
power supply (that operated on either 240vac or 12vdc) and a rack mounted
speaker. Some receivers had special housings that had storage
cubbies for four coil sets, two on each side of the receiver. Normally, the receiver was
lowest in the table rack with the speaker in the middle and the power
supply on top. The Australian Army referred to the AR7 as "Reception Set
No.1" and normally the AR7 panels were painted "army" green.
The Australian Navy used a similar version of the the AR-7.
There was also a version the had a black painted panel (may have
been the Australian Navy version) and
another version with Dutch nomenclature made for the Dutch Navy. |
||||||||
|
Marconi's Wireless
Telegraph Company, Ltd. Type C.R. 300/1 ~ VLF-LF-MW-SW Receiver ~ 15kc to 25mc The C.R. 300 was one of the "square box" Marconi/Royal Navy receivers that grew out of the original C.R. 100 versions. The C.R. 300 receivers were built mostly towards the end of WWII in 1944. Many were installed on Flower-Class Corvettes that the Royal Navy used as escorts and protection for convoy shipping. Many of the Corvettes were built in Canada and served in the Royal Canadian Navy during WWII. Additionally, some Corvettes were sold to the USA under lend-lease and operated by the US Coast Guard during WWII. The C.R. 300 had the widest continuous frequency coverage of any of the C.R.-Series receivers spanning 15kc up to 25mc. However, the price for the wide frequency coverage was the change from the double preselection used in the C.R. 100 Series to single preselection (only a single RF amplifier) and then the circuit complications caused by the need of a dual frequency IF section. There were two versions with the C.R. 300/1 having a 500kc crystal calibrator and the C.R. 300/2 having a 690kc crystal calibrator. The latter provided harmonics that fell within maritime frequency bands and there were calibration index markers on the tuning dial that indicated where those harmonics should be heard when the calibrator was on. Some versions of the C.R. 300 were built into early-1946. The circuit is an eight tube, single preselection superheterodyne that tunes continuous from 15kc up to 25mc in eight bands. The slide rule tuning dial has an illuminated "rotating drum" that shows the selected band-in-use scaling when the band switch is actuated. The two non-illuminated scales within the circular bezel are logging scales for accurate frequency reset-ability. Two different IFs are employed with 98kc used on Band 1 and Band 4. The remaining bands use a 570kc IF (see Performance section for issues related to this IF.) Each IF shielded-can contains slug-tuned transformers (permeability tuned only with fixed C) for both 98kc on the bottom and 570kc on the top. Two IF amplifiers are used in the circuit with the frequency switching occurring via the band switch. Audio output drives an internal 5" Rola speaker (3.5Z ohm VC) that's located on the right side of the panel. Two phone jacks provide 60 ohm outputs for headsets although the manual clarifies that the 60 ohms is DCR and the impedance is actually 600Z ohms. These phone jacks also provide circuit switching that turns off the panel loudspeaker when a headset is plugged in. The large left side chassis connector (PS2) provides power input along with remote loudspeaker (3.5Z separate winding) and external 600Z headset output and the "desensitizing" remote switching access pin (remote switch used to connect pin to chassis to "desensitize" the receiver.) A separate power supply is used to operate the C.R. 300. The Type 889 Supply Unit can operate on 230vac @ 50hz, +220vdc, +110vdc and +24vdc. It looks like dual vibrators on the chassis but only one vibrator is used in the circuit for the DC operation with the other vibrator being a spare unit. An efficient 0Z4 gas-filled cold-cathode rectifier is used for +250vdc B+ supply (no rectifier filament required.) The receiver's tube heaters are in series-parallel and operate on 24 volts. If 24vac is used then one side of the heater supply is tied to chassis but not in the power supply, only at the receiver power input connector (PS2.) The receiver's power requirements are 24 volts @ 0.95A and +250vdc @ 60mA. |
photo above: Marconi C.R. 300/1 receiver SN:2388 from late-WWII. These receivers were mostly used onboard RN Flower-Class Corvettes. The C.R. 300 had the widest frequency coverage of the C.R.-Series Marconi receivers at 15kc up to 25mc in eight tuning ranges. The dual tuning knobs have the rear knob with more direct "fast tuning" ratio and the front knob is the "vernier" or "slow motion" frequency adjustment. The dial lock clamps the metal skirt of the rear tuning knob. The "OFF" position between PHONE and CW will turn off the B+ but leave the tube heaters on. It was provided for "warm-up" time or for front panel actuated standby. Two phone jacks are provided. These jacks also provide switching that turns off the panel loudspeaker when a headset is plugged in. The front panel loudspeaker is a five-inch diameter Rola PM type with a very large magnet. The VC Z is 3.5 ohms and the audio output to the loudspeaker is capable of producing up to 2 watts of power. By comparison, the Phones are 600Z ohms and are driven by about 500mw of audio. NOTE: Photo taken before clean-up. |
|
Front panel controls are ON-OFF toggle switches for
POWER, AGC
and NL. The mode switch has PHONE-OFF-CW-CALIBRATE where the OFF
position turns off the receiver B+ but keeps the tube heaters on.
BANDCHANGE has eight positions. PASSBAND is W-wide, M-medium, N-narrow
and F. F is a filter position that is extremely narrow primarily for use in the LF
and VLF ranges. H.F. GAIN is the RF Gain control and L.F. GAIN is the AF Gain
control. Behind the
round cover (upper right of panel) is the "desensitizing adjustment"
(a slotted-shaft pot) which is a remotely switched-in adjustment of the RF Gain that allows reducing
the receiver sensitivity during transmitting. The desensitizing action
works against chassis-ground and is accessed at the left side power
input connector. For emergency, when all power to the vacuum tubes is no longer available, a chassis-mounted, carborundum-type crystal detector can be connected into the RF/Ant coils and the RF amplifier tube grid bypassed. The output of the crystal detector is routed to the plate of the 6V6 audio output tube which essentially connects the crystal detector to the audio output transformer. The desired signals can then be tuned on the main dial (though only using the RF/Ant coils for frequency selection.) This detector was only for emergencies and it's sensitivity was very limited. Phones were usually necessary and mode of reception was either MCW or AM Voice. There was an input shown as R.I.S. that stands for "Radar Interference Suppression" which was an external device that connected to the R.I.S. coaxial input (on the left side of the receiver chassis) and ultimately connected to the suppressor-grid of the RF amplifier tube. The R.I.S. device would block the radar pulse-rate interference if the particular ship happened to be equipped with radar. The C.R. 300/1 SN:2388 shown appears to have a Noise Limiter that is actuated with a toggle switch labeled NL ON/NL OFF. However, the switch has no connections to it and there is no circuitry in the receiver for a Noise Limiter function. The nomenclature is "stenciled" and appears to possibly be from a RN Depot upgrade that wasn't completed. The data plate apparently was removed for the NL toggle switch hole. The C.R. 300 manual doesn't mention a NL upgrade but that's not really surprising. On the "Ben Nock G4BXD - British Military Wireless Museum" website, the Marconi C.R. 300 photographed and shown there appears to have the NL upgrade installed. That particular receiver doesn't have the stenciling but has the small "on-off" tag normally found with toggle switches. That receiver also has its original data plate installed. There are other photos on the Internet showing C.R. 300 receivers with the NL circuit installed. photo left: Top of the chassis showing the tuner section on the right and the dual IF, det/agc and audio stages on the left. The shield-can and tube on the far lower left is for adjusting the 500kc calibration oscillator. The upper shield-can in the vertical row of four (has the red & blue dots) adjusts the BFO. In front of the BFO is the bandpass filter "F" assembly (has two compression trimmers.) The other five shield-cans are the dual IF transformers including the Mixer to 1st IF transformer on the Tuner chassis. The seven small chassis mounted cylinders are triple bypass capacitors. The Crystal Detector is next to the rear-most tuning condenser (Ant-RF tuning.) |
Notes on Tubes Used
- The original required tubes were,...RF Amp - ARTH2* (C.R. 300/1 version)
or KTW61 (C.R. 300/2 version,) Converter - X66 or 6K8G, IF Amp (2) - KTW61, Det/AVC - DH63 or
6Q7G, BFO - KTW61, AF Output - 6V6G, Cal Osc - KTW61. * There's some confusion (on my part) that the manual's tube list clearly shows multiple times that V1 is an ARTH2 in the C.R. 300/1 version. However, the ARTH2 is a mixer tube, a hexode-triode tube (like the X66 or 6K8,) while the manual's schematic clearly shows V1 as a pentode (like the KTW61.) My C.R. 300/1 has a KTW61 installed for V1 and it seems to work fine except the sensitivity begins to drop above 6mc (even after alignment.) The ARTH2 should work in the V1 position since the triode plate isn't connected to anything and the triode grid is tied to the hexode suppressor grid so the triode is effectively out of the circuit. I don't have an ARTH2 to try right now but when I do I'll perform a test and add the results here. ARTH2 tube ordered Oct 5, 2020 ARTH2 Update - Oct 9, 2020 - With the ARTH2 tube installed as V1 (RF Amplifier) in the C.R. 300/1 it was quickly observed that the receiver had noticeably increased reception gain, that is, louder signals. This increase in sensitivity was noticeable below about 6mc but became very obvious as the tuned frequency was increased. The usual drop in sensitivity starting around 6mc when using the KTW61 RF Amp tube didn't happen and sensitivity remained good up to 16mc (top of Band 7.) Signals were received on Band 8 but the sensitivity begins to drop off by about 18mc which is sort of expected with WWII vintage gear. With the ARTH2 RF amplifier, 40M signals are strong and 20M DX signals can be easily copied (and some 20M DX stations are even strong.) So, the ARTH2 is the correct tube as indicated in the manual and empirical testing shows that using the specified ARTH2 tube is very important for C.R. 300/1 sensitivity. Substitution Tube Notes - In most applications a 6K7G tube can be substituted for the KTW61 although there might noticeable reduction in gain. The 6K8G in place of the X66 and the 6Q7G in place of the DH63 are recommended substitutes shown in the C.R. 300 manual. The 6V6G is original equipment. The 6Q7 metal tube in this receiver was replaced with an original spec DH63 tube. Slightly increased AF gain was experienced, however, after a short time, the DH63 began to distort badly indicating the tube was probably gassy. I ended up replacing the DH63 with a glass 6Q7G tube which functioned quite well. Due to the tube socket shroud for mounting the tube shield, the 6K8 and the 6Q7/6R7 or, if used, the 6K7, must all be "G" versions (older style "G" not the "GT" versions.) Metal and GT versions of these tubes use larger bases that won't fit through the opening of the shroud. The 6V6G was standard equipment so that socket doesn't have a shroud or tube shield and a 6V6GT can be used. photo right: Under the chassis showing the RF section (upper section) with the 24 coils (3 coils per the 8 tuning ranges) for the Ant/RF stage rear, Mixer stage middle and Oscillator forward. The small cylinders are trimmer capacitors for each coil (except Band 1 Mixer and Ant/RF) and each coil L is adjusted with a slotted powdered-iron slug. Below are the IF, Det/AVC and audio output stages. The BFO and detector/AVC/1st AF amplifier stage are within the shielded compartment. The hole in the shield provides access to adjust the 98kc BFO. The date written in pencil on the cross divider is 3-10-48. This is unlikely an assembly date but might be a repair date. Most components are date coded in 1944. |
|
|
|
Refurbishment -
Started Sept 12, 2020 -
C.R. 300 SN:2388 was
in very complete and very nice condition for a shipboard receiver. It was obviously
completely neglected as far as any operation or maintenance for the past several decades.
There were six or seven components that were replacements that appeared to be
for minor repairs. These minor repairs probably took place in Canada since USA parts were used (some
of the Corvettes were built and used by the Royal Canadian Navy.) Otherwise the
receiver was original. The tube socket shroud had been removed from the
detector/AVC tube to be able to install a metal 6Q7. Luckily, the shroud
is mounted with screws and nuts. The NL ON-OFF switch was probably never
mounted. There wasn't any circuitry for a NL although the panel was
stenciled for the addition. The location of the NL switch hole required
removal of the small data plate. I added a matching style toggle switch
just to fill the hole. The Type 889 Power Unit was a disaster. Many parts were missing. Many parts were broken. The front panel was 75% surface rust. There was severe paint damage. The 0Z4 was replaced with a 5Y3GT (that substitution can't function correctly.) The power connection terminal block was hanging by a few wires and the mounting flanges were broken. No doubt, someone had spent considerable time trying to rebuild, modify, repair or do something to this Type 889 unit. Since it appears it was never really functional after the "hamstering" it never was able to power up the receiver. That was lucky since that has preserved the receiver from an amateur hack-job and that is probably why the receiver is more-or-less functional on almost all original parts. The Type 889 is covered in its own section further down this write-up. I powered up the C.R. 300/1 using a Lambda 25 adjustable B+ supply set for the +250vdc and I used a 25.6vac 1A filament transformer for the tube heaters. I powered the filament transformer with a variac so I could have some adjustability for that voltage also. I used alligator clip test leads to connect power to the receiver by way of the large left side connector, PS2. |
Tuning Dial Rebuild - The tuning shaft operates a gearbox to drive the tuning condenser. The tuning shaft is coaxial with the inner shaft being a reduction drive. A pulley is mounted to the main tuning shaft that uses dial cord to drive the slide rule dial pointer. To access all of the dial drive parts it was necessary to remove the cabinet and the front panel. With the cabinet and panel removed, then the logging dial index can be removed which then allows the logging dials to be removed. The logging dials were cleaned with Glass Plus. The dial pulley is the inner-most unit on the tuning shaft. I had to loosen the pulley to access the dial cord. I noted that the spring-load for the dial cord was totally relaxed due to the dial cord having stretched over the years. I had to add a knot as a spacer to take up the slack. Once the dial cord was the correct length again, it was then routed as needed and the pulley and dial pointer adjusted to track mechanically. Then the logging dials were remounted and aligned. The dial drum was extremely dirty but responded well to Glass Plus for cleaning. I had to add a spacer on the left side of the dial drum mount to keep it centered and in calibration (the slide rule dial resolution is vague, the logging dial scales are used for accurate frequency reset-ability.) I used light grease to lube the dial pointer track, the ball bearings in the gearbox and applied machine oil into the coaxial tuning shaft. The anti-backlash split-gears were also lubricated. The end result was the tuning was now very smooth and light with no backlash or binding. The panel was cleaned and the dial window and the logging dial window (both plexiglass) were cleaned before remounting the panel. |
|
Audio Problems
- Although SN:2388 did basically function using "test bench" power, the
loudspeaker audio level seemed to be somewhat less than expected. I had
no output on the PHONES line and the remote speaker line was also very low in
audio. I disassembled the front panel loudspeaker bracket that also doubled as a
mount for the audio output transformer in order to have easy access to
test the audio transformer windings. All audio output transformer
windings tested
okay except that when I desoldered the audio line from the transformer to the
phone jacks (and external phones,) the DCR of the transformer pin 4 to
pin 6 (chassis) changed from 1.0 ohm to around 330 ohms DC. With the
"phones line" disconnected, I powered up
the receiver and now had lots of audio on the loudspeaker line. Further isolation of the External Phones shielded cable from the Phones jacks back to the External Phones terminal on the side power connector made it obvious that there was an internal short. The shorted cable was almost entirely sleeved in yellow fabric lacquered tubing along with other wires and cables in its run from the front of the receiver chassis to the opposite side-rear. This sleeving made it impossible to remove the shorted cable without damaging the fabric tubing. I had almost decided to just leave the cable isolated and just not have an External Phones output but as I moved the shorted cable I noticed a rough hole right through the yellow sleeve and through the black rubber sleeve that was covering the cable. This ragged hole was right inline with where one of the bottom cover mounting screws would be located. Obviously at sometime in the past someone had installed a screw that was "too long" with "too sharp of an end point" when mounting the bottom cover and this screw had penetrated the cable sleeve, the cable and the center conductor causing internal damage to the cable resulting in the short. The cable damage caused the short to still be present even after the screw was removed. photo left: Inside dual IF transformer number 4 showing the upper transformer for 570kc and the lower transformer for 98kc. |
|
To repair the cable I
first had to dismount the two phone jacks to have access to the
connections. With the jacks in a good accessible position, I cut just the shield
part of the cable just above the
hole and then pulled the shield down towards the end. This
revealed the insulation damage to the center wire which I then covered with a wrap of friction
tape. Then the shield was pushed back to cover the tape and to be in contact the
other end of the cut shield. I wrapped the shield joint with 32 gauge
copper wire and then soldered the shields together. I then pushed the
black rubber
sleeve back in place covering the soldering job on the shield. I resoldered the cable wire end to the two phone
jacks and remounted them to the front panel. I then reconnected
the External Phones end of the cable to the side connector. A resistance
check now showed 330 ohms DC which is the DCR of the "phones" winding of
the audio transformer with no shorts to chassis. Other minor problems discovered while doing the audio repair were a broken wire at a junction terminal board that was from the audio section of the receiver. The AGC toggle switch had broken both rivet heads that held the switch together. The switch worked but it was beginning to slightly spread apart. I pushed the rivets back in place and soldered small wires into the rivet ends to provide a fairly strong "head" to hold the switch together. A quick power up test showed that the C.R. 300/1 now had plenty of audio. AM-BC stations only required the LF Gain to be advanced about 25%. So, the low audio problem was caused by a chassis short on the Phones line. Even though there are separate windings for the loudspeaker, the external loudspeaker and for the phones, the shorted phones winding was enough of a load on the audio transformer that it reduced the entire audio level. |
IF Alignment - IF is
adjusted for 570kc using just the top slugs for peaking. The IF for Band
1 and Band 4 is 98kc and that is adjusted using the lower slugs that are
accessed under the chassis. Although the manual suggests using an audio
output meter, this can get kind of annoying since you have to have a
loud enough signal for the audio level to show on the meter. I used the AGC
line instead since the audio gain can be kept very low. Since the RF
gain and AGC plus the signal level determine the AGC voltage level,
adjusting all alignments for peak works fine. All of the RF and IF slugs
have a small slot that requires a long (but fairly small tip) fiber blade
screwdriver. Using a steel screwdriver will add to the total L and the
adjusted
setting will change once it's withdrawn from the IF coil barrel. Although the top
adjustments are easy to access, the under-the-chassis adjustments are
through holes in the chassis and you really do need a "long" fiber
screwdriver to access these. Additionally, the slugs are coated with a
white powdery lubricant that looks terrible but works great. None of the
slugs were stuck or even hard to turn. BFO is adjusted in a similar manner in that the upper adjustment is for 571kc and the lower adjustment is for 99kc, in other words, adjust the BFO to be 1kc above the IF. However, this was the adjustment for the Royal Navy when the receiver was primarily used for CW. Nowadays, a lot of listening will be to SSB stations. It is more advantageous to set the BFO for zero beat with the IF. That way when tuning you can select USB or LSB by where you tune the signal to demodulate, that is, either above or below the signal center frequency. >>> |
>>> The Calibration Oscillator also has a slug trimmer accessed from the
top. My C.R. 300 being a "1" version, has a 500kc crystal within the
coil shield can. The easiest method of adjustment is to tune in WWV at
5mc or 10mc, switch on the Calibration Oscillator. Use N for bandwidth
and tune for peak. Then adjust the Cal slug for zero beat. RF Tracking Alignment - RF alignment per the manual uses the logging dial scales for frequency settings. This assumes that the receiver was going to have an accurate conversion table that provided the radio op with the settings for specific frequencies. The manual does caution that the slide rule dial isn't all that accurate. However, most modern users aren't going to create a logging dial to frequency conversion table except for just a few stations or net frequencies. Therefore, I just used the slide rule dial as a reference and adjusted for reasonable tracking accuracy. The top end of the band capacitive trimmers are cylinder-types and adjust very easily. The low end adjustment are the lubricated slugs. It's very easy to align the front end since each coil is clearly identified in the manual drawings. That's how I did this "quickie" alignment. I did another more accurate RF tracking alignment in late October and used the manual procedure that utilizes the logging dial for extreme accuracy. Details further down this write-up in "Workbench Revisit." |
Reworking the Type 889 Power Supply Unit |
|
The Type 889 was is in very "rough" condition. Besides its sorry
cosmetics it had been completely altered from original. All of
the wiring for DC operation was removed and several of the DC components
were removed. Both of the plug-in vibrators were gone. The
switch that selects the input voltage operation was totally
disconnected. Additional holes were cut into the front panel. The
terminal block for connecting power input had one of the mounting
flanges broken. The fuse block was completely missing. The wiring
and soldering were worse than "amateur" in quality. Workmanship
was the worst of "hamster" jobs. The mods made were non-functional (and
never would have been functional.)
Electrical Inspection - I noted that the 5Y3 (a modification) filament was connected to the same power transformer winding as the pilot lamp. That should have proved exciting if the lamp was ever changed with the power on. It was necessary to isolate most of the circuitry from the power transformer and to apply voltage to the primary of the transformer to test the which windings were going to be required. Most of the windings that were for DC operation from the vibrator had been cut, so only the AC operation windings were connected. I used a variac to apply a controlled input AC voltage level. My first connection was per the schematic and resulted in the correct secondary voltages if operation was going to be on 240vac input. I noted that the actual modified wired connections for AC input were connected to different taps on the primary than original (per the schematic.) I applied AC voltage at these connections and the resulting secondary voltages appeared to be somewhat correct with 120vac input. That's not too surprising since the receiver did come from Canada. Some pseudo-technician in the past had utilized other taps on the power transformer primary winding that were originally for DC inputs via the vibrator that now weren't being used (probably the 110vdc input.) These connections worked out for the primary to have almost the correct ratio to the secondary windings for proper voltage outputs, however, the actual voltages with 120vac input were about 30% higher than specified. Other components tested were the plug-in triple filter capacitor that tested good. The chokes tested good. The four 2uf oil-filled caps checked good. What wasn't correct was the 5Y3 filament operating on the pilot lamp winding. If operated in this manner, there's no isolation of the B+ on the 5Y3 filament winding and B+ would be floating on the pilot lamp to chassis. A better choice would be to go back to the original 0Z4 rectifier. Reworking the AC Power Supply - The solder used didn't look like SnPb (real tin-lead solder.) Either the solder was defective (some amalgam that didn't flow correctly) or the soldering iron used was too cold. In any event, the soldering for the modifications was a very poor quality with several being "cold solder joints." The same wire type was used for all of the modifications. Since nearly all of the wiring was so poorly done, I stripped out the AC portion of the chassis and started over. I used mostly NOS vintage stranded wire for the rework. I had to repair the wires from the power transformer to the rectifier plates (HV) and also the wires to the rectifier pins 2 and 7 which aren't tube connections but are tie-points for the pilot lamp wiring (as original.) Pins 2 and 7 aren't used because I wired the power supply back to using a 0Z4 cold-cathode rectifier. That way the winding only operates the pilot lamp (as original.) |
More Rework and Testing - Once the 889 was rewired it was given a "clip lead" test to see if everything functioned before the cosmetic work was started. As mentioned above, as wired and operated on 120vac, the output voltages were too high with the B+ running about +330vdc (should be +250vdc) and the heater voltage running about 33vac (should be 25vac.) Different taps on the transformer primary were measured for DCR and the next closest tap was tested but provided too low of voltages for output. However, when the power transformer was connected per the schematic for 230vac operation and that voltage level used as the input power, the output voltages were correct as specified in the manual at +250vdc for the B+ and 25vac for the tube heater supply. The only solution is to operate the C.R. 300 via the 889 on 230vac input, as designed. I have 230vac available out in the shop but not in the upstairs ham shack so I have to use a "step-up transformer" to have a 230vac power source available upstairs. Using the step-up transformer (which actually has an output of 244vac since our line is 122vac) produced +263vdc with a 65mA load for the B+. The tube heater supply was 25.9vac with no load and the pilot lamp was operating on 6.1vac. All voltages were pretty close to specs. | Cosmetics - The front panel and the cabinet both required a total repaint. Additionally, the front panel had several non-original holes that had been hacked in over the years. I was a little hesitant to strip the front panel paint because of the silk screened nomenclature but about 80% of the silk screening was already missing because of the rust so stripping the remaining paint didn't do that much damage. Fortunately, the Marconi tag was in good shape and removable. The non-original panel holes were filled with epoxy. Paint used was a Rustoleum product called "Granite Satin" which is pretty close to Marconi gray (just slightly darker.) I had to use a standard USN-style, WWII vintage fuse holder (didn't have anything like the original that was entirely missing.) I also had to use a different type of terminal block since the original one had a broken mounting flange (in fact, the terminal block used is gray rather than black.) The voltage input terminals on the block are now changed to just the 230vac input with Line-Line-Ground connections. The remaining three terminals provide monitoring access for the two lines for tube heaters and for the B+ line. When measuring the voltage outputs, for a correct reading, the receiver has to connected and operating to provide a load on the output voltages and to also provide the ground connection for the B minus line. Before painting the front panel, I metal-stamped "240 - 240 - 0 - 25 - 25 - 260" above the terminals to identify the voltage inputs with the left three being, L - L - G for 240vac input and the right three being Htr - Htr - B+ monitoring outputs. I also repositioned the Marconi data plate to be more visible above the power output connector. I also had to fabricate the bent metal piece that covers the terminal strip inputs. Reassembly completed Oct. 4, 2020. |
Power Cable Repair
- The power cable only has four wires that actually connect to the 889.
There are three other wires within the cable for remote phones, remote
loudspeaker and remote receiver desensitizing. The cable appeared to be
original with no indication that the receiver outputs were accessed by
an accessory "pig-tail" cable the contained the remote outputs.
Therefore, if the receiver outputs were needed, the end user had to make
connections to the necessary terminals on the power connector inside the
Type 889 and then route the needed wires out from there. The manual
doesn't have any recommendation of how the
remote phones, loudspeaker and desensitizing were accessed. Utilizing
the power output connector of the
889 seems
to be the easiest method (shielded cables should be used for these
remote connections.) There was some minor damage to the cable shield at one end where the shield had be cut which completely isolated that additional route for interconnecting chassis-ground between the receiver and the power supply. I had to repair the cracked insulation on two internal wires. Then wrap the wire bundle with a single layer of friction tape and then a single layer of electrician's tape. Next, a split section of RG-8U jacket was fitted over the wire bundle and then a short section of shield that was removed from some junk RG-8U was split and then fitted to make a shield "patch." The shield was soldered to each cut end of the original shield to form a continuous connection. Additionally, the connector has a tapered clamp the fits over the shield and provides a positive shield-to-connector contact. The patch was wrapped with electrician's tape to complete the repair. Power ON Test - Before connecting the 889 to the receiver, the cable was tested for continuity and for shorts. The cable connectors are identical so it doesn't matter in which direction the cable is connected to either the 889 or the receiver. Since everything had been pre-tested individually in advance, the C.R. 300-1 powered up with no issues. The measured voltages with the receiver operating and the voltages measured at the receiver were +270vdc for the B+ and 23.5vac for the tube heaters, both within spec. The receiver operated the same as it had before when running on the Lambda and the AC transformer. I half expected some sort of noise from the 0Z4 tube but the receiver's audio was normal. I checked reception from 10mc down to 19kc (MSK station HOLT) with no problems encountered. Photo right shows the 889 after rebuild. Oct 5, 2020 |
C.R. 300/1 Performance Details This section contains a lot of information on the possibility of 570kc AM-BC stations interfering with the 570kc IF of the C.R. 300/1 receiver. This is a significant problem on Band 3 while listening at night. There's a lot of information presented on Wave Traps and resonant antennas as solutions. For amateur band operation, a resonant antenna will usually be all that's required. For NDB listening a wave trap is the easiest solution. |
The 570kc AM-BC to IF Leakage Problem on Band 3 - Actually using the C.R. 300 as a receiver in a vintage military amateur radio station today will depend on where in the spectrum you intend to operate, where your QTH is and when your operation is scheduled for. While the receiver is certainly sensitive and can respond to very weak signals there is a serious underlying issue with the 570kc IF on some bands. It's quite easy for AM-BC stations operating on 570kc to be strong enough that the receiver's IF will respond to the AM station along with the tuned incoming signal. The result is a heterodyne when tuning in signals and in some cases the AM station "beats" with the reception noise and produces a constant heterodyne (and sometimes even the offending station's audio modulation.) Band 1 and Band 4 use a 98kc IF and won't have this problem but the other six tuning ranges use 570kc and might have the problem. Listening or operating during the day is not too much of a problem unless there's a strong local 570kc AM station. Nighttime propagation however allows 570kc stations from fairly distant locations to possibly produce strong enough signals that the 570kc IF would respond to them. Band 3 (200kc to 500kc) is the most seriously affected band and the closer one tunes to 500kc, the stronger the heterodyne interference becomes. Fortunately, back in the day, for shipboard installations, 500kc emergency frequency monitoring would have been accomplished using the generous "overlap" provided by Band 4 which tunes from 350kc up to 1000kc employing the 98kc IF. >>> | >>> The selectivity of the Ant-RF stage will certainly help on the higher bands but it does depend on how strong the offending 570kc AM BC station's signal is. As a test, I loosely coupled a RF signal generator to the C.R. 300 and found that I could detect a 570kc heterodyne signal up to about 10mc BUT this only simulates the effect that a very strong local AM-BC station might have. Most types of receivers that used an IF that might have operated at a frequency that also had strong RF signals present employed wave traps within the receiver input circuitry. These wave traps were adjusted to the IF and effectively nulled that frequency at the Ant-RF stage but all other frequencies would have been passed unattenuated. However, the C.R. 300 doesn't have an internal IF wave trap so the designers probably intended for the receiver to use an external wave trap that was maybe part of the ship's antenna system. It's easy to add either a series LC shunt wave trap or a parallel LC series-connected wave trap between the antenna feed line and the receiver antenna input. The wave trap usually consists of a fixed inductance and a variable C to adjust resonance. |
Band 3 AM-BC to IF Leakage Problems for NDB
Listening
- The random length, untuned end fed wire antenna is the most
susceptible to AM-BC to IF leakage since this type of antenna will
respond to almost everything in the EM spectrum. The larger the random length or untuned antenna is the more likely
that the 570kc leakage into the IF will become a problem. However,
as a rule, the further from the IF that the receiver's tuned frequency is, the more likely that the selectivity of
the receiver's Ant-RF stage will reduce the AM-BC to IF leakage to a
minimum.
Listening for NDBs will require using Band 3. Copy of NDB stations becomes more and more difficult as the tuning approaches 400kc. By 500kc the heterodyning is constant. MW DX reception only occurs at night which is also when the AM-BC propagation is best. Using just a wire antenna for Band 3 will produce rampant heterodynes during nighttime listening. I've found that the "shunt connected" series LC network wave trap works quite well at reducing the AM-BC to IF leakage to an absolute minimum when used with an end fed wire type of antenna. Another possibility (which I haven't tried) would be a hi-Q remotely tuned loop antenna that would provide the selectivity needed and probably work quite well for receiving NDBs. Additionally, installing a wave trap in the feed line of the loop would provide maximum 570kc attenuation. A shielded-magnetic loop (like the Pixel Loop) is broadband and responds to the 570kc AM stations as well as to the other MW signals. I found that the Pixel Loop was just as bad as the wire antenna for allowing heterodyne interference. A wave trap is necessary with the Pixel Loop. In my set-up, I use a 135' "T" antenna that is basically just a wire feed line. Using just the antenna for nighttime Band 3 listening, heterodynes are easily detected and become a problem by 350kc. I use a Series LC wave trap connected as a shunt (antenna feed line to chassis ground) to effectively eliminate heterodyne interference for nighttime listening for NDBs using Band 3. Using the C.R. 300/1 as a Amateur Vintage Military Radio Station Receiver - My amateur military radio operations are on 75M and are scheduled for mornings or late-afternoons. Also, there aren't any local 570kc AM stations in my area. This results in no detectible AM-BC to IF leakage interference when using the C.R. 300 as the station receiver on HF on 75M. I use a 135' tuned Inv Vee antenna fed with ladder line. No wave trap is necessary. If 630M operation is planned, luckily 472kc can easily be tuned on Band 4 where the 98kc IF eliminates the problem. As to the C.R. 300's signal reproduction, the audio response is somewhat better than communications grade especially on strong AM signals. Cabinet installation greatly improves the lower end audio reproduction. Although the top band (8) responds well to the signal generator, decent reception on this range would require an antenna with some gain. Band 7 and 20M reception would also benefit by using a yagi or quad. Note: At this time I'm using a KTW61 RF amplifier tube instead of the original ARTH2 tube. That might have an effect on the higher frequencies (Band 7 & 8) which do seem a bit insensitive when compared to Band 6 and lower. An ARTH2 tube has been ordered and when installed any performance changes will be added here. Oct. 5, 2020. UPDATE: Oct. 9, 2020 - Installing an ARTH2 tube as the RF amplifier does significantly improve sensitivity and the increase is especially apparent above 6mc. 40M signals are now strong and 20M DX signals are easy copy (and some are even strong.) Using 135' Tuned Inv-Vee antenna. The C.R. 300/1 will operate fine as a station receiver on 630M (use Band 4,) 160M, 80M and 40M ham bands. If you're using a tuned or resonant dipole for the antenna, you shouldn't experience any AM-BC to IF leakage. It's an interesting receiver to operate and virtually nobody seems to have ever heard of one being used as a ham station receiver. It's sensitive enough and the selectivity is very good. The audio response provides pleasing sounding AM signals at the panel speaker or a larger external loudspeaker can be used. I use the front panel PHONE-OFF-CW-CAL switch in OFF to turn off the B+ for standby when transmitting. The ART-13 transmitter being used has a vacuum antenna switch for receiver isolation during transmit. First "on the air" net use of the Marconi C.R. 300/1 was on Sept. 27, 2020. I didn't have the Type 889 finished so I powered the receiver using the Lambda 25 for +250vdc B+ and the 25.6vac transformer for tube heaters. The one hour long, 75M net was copied solid 100% using the C.R. 300/1. Operation of the C.R. 300/1 and Type 889 together "on the air" was the hour and a half long 75M net on Oct. 11, 2020. Copy was solid 100%. The ARTH2 RF amp tube had also been installed by that time. |
|
Oct. 29, 2020 - Workbench Revisit - After using the C.R. 300/1 as a station receiver for about one month, a few minor problems required returning the receiver to the workbench for further "tuning up." The return work was going to require some disassembly mostly involving cabinet and bottom cover removal (and front panel dismount later in the "revisit.") Installation of Tube Socket Shroud/Shield for DH63 - The first problem was mechanical because I had replaced the metal 6Q7 tube with the proper DH63 glass tube and that now required installing the correct tube socket shroud and tube shield. These parts had been removed years ago when the 6Q7 was installed. Installation was just a matter of tube socket screw removal, installing a correct style shroud and then reinstalling the screws. LF Gain Pot Problem - The second problem was a noisy AF Gain control (LF Gain) that seemed to be more intense since installing the DH63 (more info on this tube's performance, or lack of, in "DH63 Problem" below.) Access to the LF Gain pot was easy. I removed its back cover and cleaned with a DeOxit dampened Q-tip. The pot was reassembled and remounted. Band Switch Problem - Third problem was that I had to "rock" the band switch on Bands 6 and 7 to get full sensitivity. I hadn't cleaned the band switch, so that was done using a small paint brush and DeOxit. When doing the cleaning I discovered that the RF/ANT section bandswitch mounting screws were extremely loose. Both mounting screws required considerable tightening to have the switch always "inline" and to not move around when the bandswitch was operated. All of the other Bandswitch mounting screws were tight. The RF/ANT switch movement probably resulted in the misalignment in position 6 and 7 which the "rocking" of the Bandswitch knob would get the loose switch into the proper position. The IF band switch and the Passband switch were also cleaned. Re-Alignment of the RF Tracking - Since I had changed the RF amp tube from a KTW61 to the proper ARTH2, I performed a second complete RF tracking alignment. Since the installation of the ARTH2 improved the high frequency sensitivity so much I thought that the tube must have slightly different characteristics and a RF tracking alignment might show more improvement. Also, the first alignment was just a "quickie" - kind of a "roughed-in" alignment to make sure everything worked okay. I had used the slide rule frequency dial settings for tracking and that dial doesn't have any accuracy at all (even the manual states that the slide rule frequency dial is just for "reference.") The manual provides exact frequencies that correlate to 23.00 on the logging dial for the low end and 1.00 on the high end for each of the eight bands. To align the tracking accurately one has to use the logging dial since that is mechanically driven along with the tuning condenser. One has to use a 200pf dummy antenna load between the receiver and the signal generator on Bands 1, 2, 3 and 4. On Bands 5, 6, 7 and 8 a 100 ohm CC resistor is used for the dummy antenna load. The tracking adjustments were off by quite a bit, probably because of the way I did the initial alignment (a "quickie" alignment.) After alignment, I've noticed that the relationship of the logging dial/f versus the slide rule dial indication is that the upper and lower ends of the slide rule dial are off at the specified logging dial numbers but the majority of the slide rule dial indication is relatively accurate. RF/Ant Coil #8 Problem - Another problem found while doing the alignment was that the RF/ANT adjustments wouldn't work on BAND 8. I found that one of the terminals on coil #8 had three different wires "wrapped and stacked" but the bottom wire in the stack wasn't soldered. I cleaned the non-soldered wire and then flowed new solder on the entire joint. Afterwards BAND 8 RF/ANT adjustments worked and could be aligned (and they were.) This improved BAND 8 sensitivity noticeably since before the RF/ANT coil wasn't "tuned." It's probable that this connection had been unsoldered since the receiver was built and the function of BAND 8 had relied on the wrapped mechanical connection but with decades of minor oxidation and dirt accumulation that mechanical connection had become non-conductive. HF Gain Pot Problem - The HF Gain pot had been giving erratic contact at the the top end of travel. Since the LF Gain pot cleaning had turned out so well I decided to do the same to the HF Gain pot. After dismounting the pot and removing the back cover an inspection revealed some surface cracking on the carbon element near end of travel. I cleaned the element and reinstalled the pot then set-up a temporary operation of the receiver which showed that the HF Gain still had a problem. Replacement of the HF Gain pot was unfortunately necessary. The junk box provided a very close look-alike 10K pot that was installed which corrected the HF Gain adjustment problems. The Front panel had to be dismounted for the HF Gain pot work so at this time I also lubed the IF switch cam with viscous grease. DH63 Problem - It was noted that when the receiver was connected to the station antenna the CW sensitivity seemed okay but when the BFO was switched off for AM reception the background noise totally disappeared and only very strong signals could be heard and they were badly distorted. This problem had developed while the receiver was on the workbench and it seemed to progressively get worse each time the receiver was operated. While I had thought the DH63 was a good tube (after all, it was sold as "used-tested good,") I replaced the DH63 with a glass 6Q7G. With the 6Q7G installed, the AM reception then had the usual background noise and sensitivity in both CW or AM was "back to normal." DH63 tubes have to be purchased from the UK and all examples that I've seen for sale are in "used-tested good" condition. I'm sure that this DH63 "tests good" but tube operation in a tube tester is usually just long enough to perform the test and that's all,...maybe one minute,...tops. It's quite different from the tube's operational performance in a receiver where the circuit parameters are quite different and length of time "powered up" might be measured in hours. For now, the 6Q7G will do fine. It looks absolutely correct since the "G" version fits through the tube socket shroud, it has the original style tube shield installed and most important - the 6Q7G is the recommended substitute for the DH63 in the receiver manual. 570kc IF Alignment versus Bandwidth Performance - In the copy of the C.R. 300 manual that I have the "Receiver Alignment - page (18)" is missing. This page details the IF alignment procedure. I had to experiment with some different adjustments to see what worked best. The best performance for the IF bandwidth function requires accurately aligning the 570kc IF with the receiver in the "N" Bandwidth position. The IF circuit boosts the gain when in "N" to compensate for the narrow bandwidth. If the 570kc IF is aligned in either the W or the M position then the gain boost for N is not apparent and the bandwidth doesn't seem to widen or narrow per the Bandwidth selection. The 570kc IF accurately aligned in the "N" position gives the user a wide bandwidth response in W, slightly narrower bandwidth in M and very narrow bandwidth but with boosted gain in N. This works out very well for CW, which was the commonly used mode during WWII. Since only a stronger AM Voice signal audio bandwidth would benefit in the M or W position, the slight reduction in IF gain is expected (but not really all that apparent) when these types of signals are tuned. Workbench revisit completed on November 6, 2020 and the C.R. 300/1 was returned to the operating position paired with the Navy ART-13 transmitter. Combo used for Nov. 7th - MRCG net on 3.985mc with no problems. This combo has been used many times since then and the receiver is an excellent performer on 75M Phone. I've also listened to the various Chinese and Korean Frequency Marker Beacons on several different frequencies. Also, Gander Air on 10mc and Trenton Military on 15mc, both aviation weather stations. So, the C.R. 300/1 is a good, dependable receiver that is able to provide reception many different types of transmissions. |
|
|||||||||||||
Re-Restoration - I obtained CSR-5 SN:394 around 2010 from fellow ham and collector NU6AM. The receiver had been cosmetically restored earlier by K6DGH, who had had the cabinet and panel powder-coated in a brownish-gray color that was substantially darker than the beige wrinkle that has been found on many CSR-5 receivers. NU6AM, Jim, had obtained the receiver in "unfinished" condition and needing some electronic work. Jim did a "re-cap" job that was not particularly sympathetic using "yellow jackets" and IC electrolytics. I believe that Jim wasn't too impressed with the finished CSR-5 performance and that's why he eventually sold it to me. At that time, I repainted the dial escutcheon black wrinkle finish because it had been flat black with white rub-on "block" letters that weren't original (and were obviously an amateur application.) I also made new cables for the power supply to receiver hook-up since the cables that had been used were made from very small gauge wires that didn't look sufficient to provide the power supply connections without a substantial voltage drop. I didn't do anything else to the receiver except use it for a short time and then put it on display in the Western Historic Radio Museum in Virginia City, Nevada. Fast-forward to 2020,...I had closed WHRM in 2012 and moved to Dayton, Nevada. The CSR-5 had been stored from the past seven and a half years in my "cool room" in the shop in Dayton (a well-insulated storage room where the temp extremes are from 30F to maybe 75F.) I had just recently gone through a Marconi R1155 and its impressive multi-colored arc'd dial got me interested in that large multi-colored dial on the Canadian Marconi receiver. I had to extract it from the "cool room" and bring it to the upstairs radio repair location to take a better look at it and reacquaint myself with how it performed. The CSR-5 still worked as it had,...okay, but seeming like it could do much better. The tubes tested okay but the band switch was very erratic and needed cleaning with DeOxit and a small brush. All controls were given the DeOxit treatment. The split-gears in the tuning gear box were stuck together so a flush with WD-40 loosened them and that eliminated the backlash. The BFO and NL toggle switches seemed non-functional but a "spray down the barrel" with DeOxit cleared up the internal oxidation that was causing the problem. I was sure that neither of the former owners had performed an alignment. If a full IF/RF alignment hasn't been performed, doing one will certainly improve performance significantly. The IF adjustments and RF L adjustments all have lock-nuts so be sure to loosen when adjusting and snug-up when finished (while watching the output meter to verify that the adjustment doesn't change with lock-nut tightening.) With this CSR-5, all of the alignment adjustments were far enough out that quite a bit of improvement was gained by the procedure, especially the tracking which now easily meets the 0.5% accuracy tolerance specification.
|
WWII Navy Gear - Part 1 WWII USAAF, USCG Comm Gear - Part 2 Home Index
|
Radio Boulevard
Western Historic Radio Museum
Vintage Radio Communication Equipment Rebuilding & Restoration Articles,
Vintage Radio History and WHRM Radio Photo Galleries
1909 - 1969
- 60 years of Radio Technology -
This website created and maintained by: Henry Rogers - Radio Boulevard, Western Historic Radio Museum © 1997/2023