|
Radio
Boulevard
|
|
|
The Twenties and Before |
|
Radiomarine Corporation of America In 1907, Wireless Specialty Apparatus Company (WSA) was formed by Greenleaf Pickard who had cataloged over 1500 minerals and combinations that worked as detectors in 1903. Pickard also patented the Crystal Detector. Also, involved in the formation of WSA was Col. John Firth and his patent attorney, P. Farnsworth. John Firth had worked with Reginald Fessenden's NESCO (National Electrical Supply Company) at the same time and, probably because of this association, all WSA model numbers have the "IP" prefix which meant "Interference Preventer" - a Fessenden term for "selective tuning." United Fruit Company was an early user of WSA gear on their ships and in their shore stations. United Fruit Company had been somewhat involved in the radio business since 1903 and, in 1911, they bought WSA. Later, in 1913, United Fruit formed another radio company, Tropical Radio Telegraph Company, mainly for setting up and operating communications stations around their extensive property holdings in Central and South America. United Fruit was aware that their purchase of WSA gave them all rights to all Crystal Detector patents that WSA had through Greenleaf Pickard's association. The crystal detector dominated wireless receiver operation, especially through WWI. Right after WWI, the crystal detector patents were still of major importance and because of this patent, United Fruit Company was invited into the new cross-licensing agreement that was headed by General Electric and included Radio Corporation of America (GE's creation, formed in November 1919) along with AT&T (vacuum tube patents) and Westinghouse (Regenerative Detector and Superheterodyne patents and International Radio Telegraph Co.) With the cross-licensing agreement, these companies shared patents and products toward their advantage and tried to exclude most other companies and their competition by threats of lawsuits. WSA had several advertisements published that informed everyone that they owned the Crystal Detector patent and were prepared to sue any infringing companies. WSA had built the SE-1420 receiver for military contracts at the end of WWI along with other military wireless equipment during WWI. Now, the cross-licensing allowed UF/WSA to produce the SE-1420 for commercial users and have them sold by RCA. (RCA hadn't created a division specifically to handle all maritime equipment and operate coastal communications stations - yet. For the first half of the 1920s, RCA was considered essentially a sales agency for radios produced by the other members of the GE-Westinghouse-RCA-AT&T-WSA "Radio Group.") For the new commercial market, the designation of the military SE-1420 was changed to IP-501. A matching two-stage audio amplifier was now available, the Triode Type-B Amplifier. A Long Wave Adaptor was also available for operation at VLF frequencies. The IP-501 equipment was primarily for commercial shipboard use. |
|
|
Radio Corporation of America In 1919, GE was working on a sale of their important wireless patents, including the Alexanderson Alternator, to the British Marconi Company (Marconi's Wireless Telegraph Company.) When the Navy "got wind" of this intended sale, they were livid. Especially since the information had come to them through "ordinary trade channels." The Navy had just spent the past few years (WWI) securing and protecting US wireless stations along with protecting the transatlantic communication cables which were the sole communications between the US and Britain during WWI. Developing reliable maritime radio communications was a top priority for the Navy and, here was GE, ready to sell the major wireless licenses and patents to the Brits. The Navy sent their top brass, headed by Commander Stanford Hooper and Rear-Admiral William H.G. Bullard, to GE to persuade them to not sell USA wireless patents to a foreign nation, even if they were a longtime ally. The Navy wanted GE to create a "radio company" that would "do everything in radio." That would include building equipment for the Navy, operating and maintaining commercial wireless stations, building and selling radios and parts and, generally, be the exclusive manufacturer of "all things radio" for the United States and especially for the Navy. Admiral Bullard, who was the newly appointed Director of Naval Communications for the Navy Department, was able to convince GE's chief console and vice-president Owen Young that this new "radio company" was not only a patriotic response to British Marconi's intent to purchase the USA wireless technology but that such a "radio company" could be a very lucrative arrangement for the GE. Owen Young convinced the other GE board members that the sale to British Marconi should be cancelled. In October 1919, GE set up the preliminary version of the "Radio Corporation" using just their own assets. This didn't seem to be enough since no manufacturing facilities were included, especially for the type of company the Navy (and now GE) envisioned. GE knew that the American Marconi Company's management had long been concerned that the majority of their ownership was in British hands. Ed Nally Jr, vice-president of American Marconi and A. G. Davis (of GE) were sent to Britain to first, cancel the GE-British Marconi deal and to second, negotiate a new deal to buy American Marconi. Davis and Nally were successful and were able to have GE buy the entire American Marconi company, including their manufacturing plant in Adelaide, New Jersey. In November 1919, RCA was officially formed and announced to the wireless world. Ed Nally Jr. was the new president of RCA with Owen Young becoming a powerful RCA board member. The formation seemed set to launch the "Radio Corporation" into the wireless equipment manufacturing business, but, at the last moment, General Electric decided that too much strategic economic power would be in the hands of RCA and that GE would have little control of RCA operations. To have and maintain control over RCA, GE retained the American Marconi plant for themselves and essentially took all of RCA's ability to function as a potential manufacturer away from them. RCA became what was essentially a sales agent for all of the members of the "Radio Group," an unofficial name for the patent-sharing and cross-licensed radio companies headed by GE.However, all of the old American Marconi responsibilities became new RCA duties. The old Marconi Institute for training radiomen became the RCA Institute performing the same function. Many of the American Marconi commercial coastal wireless stations now had to be maintained by RCA and, in some cases, operated by RCA. However, RCA was determined to become more than just a sales-agent for the Radio Group and throughout the 1920s was slowly purchasing various patents, developing the radio business and growing into what the Navy had first envisioned. |
|
|
RCA's 1920s Maritime Radio Business,... Since RCA was formed from the American Marconi Company, maritime radio was the major part of the company's business. RCA was created at the instigation of the Navy for their maritime equipment needs but the commercial ships were also being supplied with Marconi gear from American Marconi and now that end of the maritime business was RCA's responsibility. GE produced and rebuilt some Marconi receivers around 1920 but this work was performed at the old American Marconi plant in Adelaide, New Jersey. This was certainly because RCA had no manufacturing ability at the time. In a short time an arrangement was made to have the maritime equipment built by another member of the "Radio Group" with maritime radio experience - Wireless Specialty Apparatus Company. The general public had become interested in the new Radio Broadcasting by 1921 and the market for entertainment radios sky-rocketed. It was a lucrative market that seemed endless in the demand for broadcast radios so, naturally, most manufacturing concentrated on radios for broadcast reception. The commercial maritime users were certainly not as numerous as the general public radio enthusiast market and because of this WSA's production level was never really very high. WSA built some home radios for RCA to sell, the AR1375, for example, but their main production was maritime radio equipment. RCA handled the sales of IP-501 and IP-501-A receivers and other wireless accessories from 1920 until sometime in 1923. By 1924, the name "Wireless Specialty Apparatus" disappeared from the scene, apparently purchased or in some way absorbed by RCA and possibly by "Radio Group" pressure. It was certainly to the advantage of RCA to have a maritime radio manufacturing business added to their other maritime radio responsibilities. This RCA ownership now forced United Fruit to depart as a radio manufacturing business owner and concentrate on their fruit and shipping business. RCA assumed the ownership of the Crystal Detector patents from WSA. In order to be able to build maritime radios RCA also acquired the WSA manufacturing plant in Boston. From 1924 up to sometime in 1927, RCA apparently carried on managing the building of the receivers using the WSA plant and components then selling them as commercial marine receivers. The other members of the "Radio Group" apparently weren't interested in the marine radio business since they were concentrating on the mega-profits from the Radio Broadcasting and Entertainment side of RADIO. |
In 1922, the IP-501-A was introduced. It combined the IP-501 and the Triode Type-B Audio Amplifier into one long cabinet. The consolidation made the IP-501-A easier to build since everything necessary was in one cabinet rather than two separate units as in the IP-501 and Triode Type B amp. From about 1923 to 1927, IP-501-A receivers were built at the WSA plant and were marked as "Radio Corporation of America" products. RCA were still using WSA parts (Faradon was the WSA trade name for components like condensers that are found in later Radiomarine equipment well into the 1930s.) |
|
Radiomarine
Corporation of America
is finally born as a division of RCA,... In 1927, RCA acquired Independent Wireless Company and it was combined with Wireless Specialty Apparatus (owned and operated by RCA so the WSA name wasn't used after 1924) to form Radiomarine Corporation of America. Independent Wireless' Chas. Pennill became Radiomarine's Vice-President and J.P. Duffy came over from RCA to act as Superintendent of Production. With the formation of RMCA, all RCA marine radio operations and sales were then transferred to Radiomarine Corporation of America as a division of RCA (later called "a service of RCA.") RCA had been operating many of the coastal wireless stations and providing "Radiogram" communications and those responsibilities were passed on to Radiomarine Corporation. James G. Harbord was president of RCA at this time (1922 to 1930) but General Manager David Sarnoff ran the "day to day" operations of the company out of the New York City office. By the late twenties, the IP-501 family of receivers were beginning to show their age. The later versions eliminated some of the unnecessary or expensive circuits and parts but the basic style was still easy to recognize. Many receivers were still in use well into the late thirties. Their operation was easy, they were reliable and their excellent performance in the hands of experienced radio ops had resulted in the receiver becoming the "standard" shipboard receiver for at least a decade and a half. Advertising for RMCA was still using photographs of shipboard installations showing the IP-501-A as late as the post-WWII time period. The selling price of the IP-501-A was $550 in 1922 (listed in "Radio Enters the Home," a catalog of products available through RCA though built by WSA at this time.) By March 1930, RCA had acquired most of the patents and power that had formerly been shared by the GE-Westinghouse-AT&T "Radio Group" as part of a settlement of an Anti-trust suit brought by the government. That resulted in the RCA becoming the omnipotent corporation that controlled almost everything in RADIO in the thirties and forties. The Broadcast and Entertainment end of the radio business was handled by the Victor Division of RCA, otherwise known as RCA-Victor. Other divisions handled other parts of RCA business, like RCA Manufacturing Co., that generally built all military contracts and commercial radio equipment. Radiomarine was the division that handled everything in the maritime radio business. It's possible that by the late-twenties, after RCA had acquired the old Victor Talking Machine Company manufacturing business and created RCA-Victor, that Radiomarine equipment could have been built in that huge facility in Camden, New Jersey, but Radiomarine Corporation continued to use "New York, N.Y." (RCA corporate headquarters) as their location. By the thirties, RMCA was running all of the RCA coastal marine stations that provided Radiogram service to ships at sea. RMCA built (probably at RCA) marine radio gear that was installed on commercial ships. During WWII, RMCA supplied the Liberty ships with radio gear. Around the mid-1960s, RMCA was eliminated as a division (or service) and RCA handled all of the maritime business directly - at least, what was left of it. United Fruit Company went on to political involvement in some Central and South American countries in an effort to protect and/or enhance their extensive banana and pineapple properties. Eventually, United Fruit Company merged with other fruit companies to become United Brands (in 1970) and then was reorganized in 1984 as Chiquita Brands International. |
|
|
Radiomarine Corporation's The receiver shown in these two photographs is a late-version, RMCA-built IP-501-A, certainly from the late-twenties. The photo left is a close up of the ID tag of an IP-501-A showing Radiomarine Corporation of America as the builder of the receiver. The photo right shows this beautiful example of the late-version of the IP-501-A. This receiver is probably built between 1927 to 1930 (or maybe even a bit later.) Note that the telephone condenser switch was eliminated and bakelite capped binding posts are used on these later receivers. both of these great photos are from Grant Kornberg |
|
|
Restoration of a WSA-RCA IP-501-A Receiver |
|
Description of the WSA-RCA Version - The IP-501-A was the classic shipboard wireless receiver. Its incredible performance, robust construction and high reliability made it the "standard" for all maritime receivers that followed for the next decade (even longer for commercial ships.) It was developed from the Navy SE-1420 and later IP-501 series of wireless receivers. A three-circuit regenerative/autodyne detector is combined with a two-stage audio amplifier along with all the extra features that would be required for reliable communications at sea. Wavelength coverage is from 300 meters up to 7500 meters (1000kc down to 40kc, though most receivers tune from 1200kc down to 37.5kc.) The oak cabinet along with the back of the front panel are lined with copper sheet and and an extra copper shield isolates the Antenna Tuner from the Secondary Tuner for top-notch selectivity. Construction is heavy-duty, quality is first-rate and the performance is incredible. All screw and nut joints are soldered after tightening at assembly. This was to prevent anything from coming loose inside the receiver with the constant vibration encountered onboard ships at sea. The solder also protected the screw threads from saltwater-air corrosion. The IP-501-A receivers (and their variations) continued to be built up into the late-twenties, perhaps even later. Manuals were available from RMCA as late as 1936. Most IP-501-A receivers were removed from ships just prior to and during WWII since their regenerative detectors easily coupled into the antenna tuner section and then up to the ship's antenna. An oscillating IP-501-A (necessary for CW reception) could be easily received as far away as five miles when operating at sea. The Navy had new specifications (implemented just before WWII) that a receiver onboard a Navy ship couldn't produce in excess of 400 pico-watts of oscillator RF energy on the ship's antenna. This was a specification that directly affected ALL non-isolated regenerative detector receivers AND it also included the high frequency oscillators in superheterodyne receivers. The Navy didn't want mutual interference between the multiple receivers and other equipment onboard their ships. Additionally, there was a genuine concern that enemy ships could pick-up the radiated RF energy with sensitive equipment and could then use direct-finding techniques and sensitive equipment to locate Navy ships. Restoration of the IP-501-A - Back in 1979, I bought this IP-501-A from a ham friend (W7IND) who had traded a telephone pole for it. The price of $75 took into consideration the substantial amount of work that was going to be required to restore the set. >>> |
|
|
|
>>> Back then, with no Internet, all parts had to be advertised for in Radio Age magazine with the hope that one of the many collector-readers might have what you needed and would be willing to sell the part. It wasn't unusual for part locating to take a year or more. The IP-501-A was missing almost all of the buss wiring, it had the incorrect AF transformers, the Telephone Switch was gone along with the correct tube socket assembly. Additionally, the set's front panel was in pretty rough condition as were all of the nickel-plated hardware parts. The cabinet was actually in pretty good condition and the parts that remained were also in good condition. My first restoration (1979) just got everything cleaned up and looking pretty good but I didn't want the IP-501-A to be just a "shelf queen" so I began to advertise for specific parts I needed to get it operational. That resulted in the second restoration the goal of which was to have the IP-501-A functional. I used black plastic tubing for the "spaghetti" on the new buss wiring and the interstage transformers were two old Jefferson transformers. I had to make several parts. I made the tube socket assembly from .25" thick original vintage bakelite. The nickel-plated tube socket shells came from an old TRF battery set. One small knob had to be cast in epoxy resin with black filler so it would come out looking like hard rubber. The metal parts that had to be made were two binding posts and one knob pointer. I got the IP-501-A running in 1982 and eventually wrote an article on it for Radio Age (April 1984 issue.) Don Patterson (editor for Radio Age) sent me some material on the IP-501-A that included an article written on the IP-501-A receiver back in the 1960s for Popular Electronics. The article had a good photograph of the inside of an original IP-501-A, something I had never seen before (remember, this was long before the Internet.) My second restoration had been incorrect in several places and needed to be redone to be an accurate representative of the IP-501-A, both physically and operationally. I started to search for more parts for the new restoration. The easiest to find were the two RCA audio interstage transformers. The new buss wire used was now 12 gauge (I had used 14ga. before) which also was easy to find. The lacquered tubing or "spaghetti" was a reproduction product supplied by one of the few dealers then around and doing business. It looked very close to original. >>> |
|
>>> The Condenser Switch and assembly was rebuilt from correct vintage parts and then installed in a painted black, brass metal box I made. All of the hardware on the front panel was removed and then cleaned and re-nickel plated. The tuning dials were listed originally as "German Silver" but that is really just a fancy name for heavy nickel plating with a little bit of copper in the nickel alloy. In order to have all of the hardware parts match, since some were new repros, I had to re-nickel plate all of the hardware. That way it all matches and looks the same vintage. Disassembly is difficult because all screw and nut assemblies are soldered. The technique for removal is to use the soldering iron to heat the solder and nut, then while the joint is still hot use a nut driver to remove the nut. While hot, the solder isn't very strong and the nut will back off with very little effort. I cleaned the screws and nuts before reassembly for better solder flow. This last restoration was finished in late-1984 and resulted in the IP-501-A looking very original and functioning as it should. Or did it? I found that I could not get the receiver to oscillate below about 250kc no matter where the TICKLER was set. Reversing the variometer leads got the receiver to oscillate correctly but the wiring layout didn't match photos of original IP-501-A receivers. I finally revised the way I had done the buss wiring which then allowed the variometer stator input to connect to the plate and the front of the variometer rotor to be the output. This got the IP-501-A looking correct inside and working the way it should - able to regenerate down as low as the receiver would tune. For the ultimate information source on the SE-1420, IP-501 and IP-501A wireless receivers that includes construction, restoration and operation of these marvelous receivers and also "Tuning in NDBs with the IP-501A" includes a NDB log of stations received (103 stations in 3 weeks time - well, that was in 2009 when there were a lot more NDBs transmitting than now in 2025.) See "SE-1420, IP-501 & IP-501-A - The Classic Shipboard Wireless Receivers" - use the Home Index link at the end of this write-up. |
|
|
The Later-Thirties and the WWII-era |
|
Radiomarine Corporation of America
AR-8503
MW, LF & VLF Radio Receiver Introduced in 1937
|
|
|
The AR-8503 was a commercial VLF, LF and MW receiver that was introduced in 1937. Initially, only the four-tube regenerative receiver was available but as WWII shipboard requirements changed, RMCA provided the AR-8503-P Pre-selector to reduce the regenerative detector leakage to the ship's antenna. Additionally, most commercial ships in the late-thirties operated on DC voltage, so the AR-8503 usually was operated on the ship's DC power set-up with dry cell batteries supplying the B+ voltages. However, later a RM-6 rectifier power unit was provided that allowed operation on a ship's AC power (most USN ships operated their radio gear on AC.) This write-up section mainly covers the refurbishment and performance of AR-8503 SN:42590. |
Mar 7, 2022 - Inspection - Today, I was given a disassembled AR-8503 as payment for a small testing job I did for Ham & Hi Fi (Sparks, NV.) The serial number is 42590 and that probably indicates the receiver was built in 1942 (the first two digits usually indicate the year of manufacture for RMCA equipment.) The receiver is about 95% complete but is missing (as always) the small round RMCA knobs although it does have the National Co. "N" dial present. The four shock mounts are gone and the six binding posts are also missing. The AR-8503 had been disassembled a long time ago and most of the parts were supposed to be in two small fruit juice cans inside the cabinet but, with all the recent moving around, the cans had tipped over and screws, washers and small parts were all spilled into the cabinet (BUT, at least they were present.) The AR-8503 was introduced in 1937 but this receiver was almost certainly built in the early part of WWII for commercial shipboard use on relatively small vessels. I've never been able to find any of the small RMCA knobs in any junk boxes - ever! But, at a little over 1" diameter, the round Hallicrafters knobs that were used in the late-thirties up through WWII look fairly close to the RMCA knobs and will have to do as a suitable substitute. The Halli knobs don't have the flutes that the RMCA knobs and they lack the metal pointers that the two bandswitch knobs had but they're still about the closest-looking, easily-found type of knob. I didn't have six "matching" EBY binding posts. Nearly all of the large quantity of vintage EBY binding posts I found in searching the junk boxes had engraved nomenclature on top of the thumbnut to indicate their intended use. I used six EBYs that were in good condition and then removed the engraving and polished the top of the thumbnut to end up with six non-engraved EBY binding posts. The rotating latch was missing from the tube inspection door but it's easily made after finding a suitable metal thumb screw (found in the vintage screw boxes.) There wasn't a pre-selector with this AR-8503 but many of these receivers were originally used on commercial ships without a pre-selector (the pre-selector was actually designed a few years after the AR-8503 was introduced.) The receiver is sensitive enough by itself (phones will be necessary) and the pre-selector was mainly for antenna isolation from the regenerative detector - a mandatory requirement on Navy ships. Also, no AC power supply but many commercial shipboard AR-8503 receivers ran on the ship's battery-generator DC supply for tube filaments and dry cell batteries for the B+ requirements. Voltage requirements are +6vdc, +22vdc and +90vdc. |
|
| Mar 12, 2022 - Front Panel Clean-up - It can't be seen very well in the "before" photo but the front panel appeared to be stained. I tried using lacquer thinner and this made the staining really bad (I did that on Mar 10.) That made me think that the panel had probably been sprayed with a "clear coat" of some type and that's what was causing the "stained look." I wondered whether it might have been MFP but the coating didn't have the characteristic yellowish tint so I thought it was something like Krylon Clear lacquer (a check of the National "N" dial confirmed that the coating was something like Krylon and not MFP.) To remove this "modern" coating I went to the old standby, NaOH sodium hydroxide, the main component of Easy Off Oven Cleaner. This is basically lye, that is, a caustic solution that is good for removing some types of stains (especially organic types.) The EOOC got rid of about 80% of the staining (Mar 11.) I went back to the idea of "clear coat lacquer" and tried heavy-duty paint stripper (non-methylene-choloride type.) That removed all of the panel problems that had looked like stains and left the textured panel looking fabulous. |
|
|
Mar 13, 2022 - Nomenclature Engraving Paint Fill - The only "fallout" from the aggressive panel cleaning was that most of the paint used for the nomenclature fill also got "removed" in the process. Although it's tedious work, it's not difficult to do the engraved nomenclature fill using Artist's Acrylic paint in Mars Black. I apply a small amount of paint using a Q-tip and work the paint into the engraving. I let the paint dry for a couple of minutes. Then using a small, folded paper towel piece dampened with Glass Plus I carefully remove the excess fill paint which leaves the engraving paint fill where it belongs (it takes some practice to do this efficiently but it's very easy to learn and the results look original.) The "arrows" that are on four of the control nomenclature engravings require dabbing a small amount of paint (using a tooth pick) into the "arrow" recess. The damp paper towel has to be wrapped tight over a straight edge (I used a razor blade edge) to prevent the towel from pulling the paint out of the "arrow" recess. I used Glass Plus applied with a small short bristle brush for clean up of other acrylic paint goofs or excess paint that had gotten into the texturing of the front panel. It works great as long as the paint hasn't dried completely. Once the engraved nomenclature paint fill is thoroughly dry (overnight is enough) the panel can be rubbed down with a soft cotton cloth. |
| Mar 14, 2022 - Reassembly - This was accomplished without any problems since I had previously sorted all of the various types of fasteners and insulators that had been in the fruit juice cans. The front panel was mounted first using the seven 8-32 OH dress screws. The front panel is also secured using the mounting nuts for the various controls. The COUPLING air variable has to be mounted with shouldered insulating washers since it's in series between the antenna and the tuning condenser. Also, the two phone jacks have to be insulated using fiber washers and shouldered mounting nuts. There isn't a flexible coupler between the "N" dial and the tuning condenser so it has to be centered and connected to the tuning condenser shaft before the dial base mounting screws are tightened. The "N" dial didn't have the internal mounting screws installed but the proper screws were among the sorted fasteners. I mechanically set the "N" dial to "0" with the tuning condenser at full mesh. The remaining controls were mounted in the normal fashion with dress washer and mounting nut. The remaining knobs were then temporarily installed. | Quick Testing - I hadn't modified the EBY binding posts yet so the voltages required were connected to the lugs that would normally mount to the proper binding post. I used test clip-leads for the connections. For a power supply I used the CRM-20096 from the RAZ-1 set. The AR-8503 receiver was out of its cabinet for the testing. The initial test used the 162' EFW antenna and the reproducers were Hi-Z TRIMM phones. I started with Band 4 at the AM-BC band and tuned KPLY on 630kc, a local (Reno, NV) station with a very loud signal. A few other AM-BC stations were also heard just below KPLY but since it was about 4PM in the afternoon they weren't DX stations. Next, I switched to Band 2 and tuned around for WWVB 60kc which should have been at 58 on the dial - it was on 57. But, the WWVB signal is pretty broad so it was easy to find and it was a strong signal. Next, I switched to Band 1 and looked for NLK 25.8kc from Jim Creek, WA and it was coming in strong. So was NPM 21.8kc from Hawaii. Both of these USN MSK stations were tuned in very close to the dial settings shown in the manual. For daytime listening, the AR-8503 seemed to be working fine and I was surprised at the receiver's sensitivity when used without a pre-selector. This AR-8503 seems to do a pretty good job just by itself. But, more evening testing will be required. I'll probably use the EFW antenna first and then compare those results with another reception test using the Pixel Loop antenna. |
|
Mar 14, 2022 Reception Test - 2200hrs to 2220hrs PDT MW 408kc - Moses Lake, WA Using 162' EFW antenna - Hi-Z TRIMM phones |
Mar 15, 2022 -
Refurbishing the "N" Dial - The "N" dial had been
sprayed with some type of clear coat lacquer (not MFP) and some
of the lacquer was laid-on so thick it was on the verge of
running (there were even bubbles in the coating.) I disassembled
the dial to prevent the stripper from getting into the vernier
mechanism. I used heavy-duty stripper (not methelyne-chloride)
and that removed the lacquer coating easily. The index scale and
the upper and lower part of the base assembly also needed to
have the spray lacquer removed. The entire stripping process
only took about 15 minutes to complete. The dial pieces were
washed in water and then washed again with Glass Plus. After
drying, the "N" dial was reassembled and remounted.
Hallicrafters Knobs as RMCA Substitutes - As mentioned, these knobs that were used on a lot of Hallicrafters equipment from the late-thirties through WWII are a pretty good match as substitutes for the missing RMCA knobs. They have a similar design and are only slightly smaller than RMCA knobs. One thing I needed to do was to make a white index "dot" on the knobs like the RMCA knobs would have. This required drilling a shallow hole towards the rim of each knob and filling the hole with off-white paint. White is usually way too bright so I mix in a little light brown and a small dab of black to end up with a manila-color for the fill that will look "aged." The two RMCA bandswitching knobs had metal pointers that would have been difficult to make and since these were only "substitute knobs" I just went with the "dot" indicator to match the other Halli knobs. |
|
|
License Tag - The "license" tag that mounts on the top of the cabinet and definitely ties the AR-8503 to commercial shipboard use with its reference to "contract of sale" in the wording. The tag was in a slightly oxidized and dirty condition. I have good cleaning results by performing a very light "going-over" the tag with 0000 steel wool and WD-40. Light pressure is all that's required to clean the oxidation and leave the lettering in perfect condition. The tag was bent and needed some bodywork to straighten. Photo to the left shows the cleaned tag. |
| Cabinet
Cleaning, Bodywork, Painting and Shock Mounts - I
used Glass Plus and brass brushes to clean the wrinkle finish
paint. The top of the cabinet was bent and needed a little
bodywork to straighten. The exterior paint was in fair condition
so all that was needed was to do a "wash." For this process, I
mix Artist's Acrylic to match the original cabinet color and
then thin that paint mix with water until it's about as thick as
a thin milk shake. I then use a sponge to apply the paint in a
dabbing technique. Also, some touch-up was required before the
wash to cover a few of the larger bare metal spots. The shock mounts were threaded 1/4-28 so mounting only required machine bolts and lock washers. These mounts were olive drab and the missing originals were silver but I painted them grey with the same touch-up paint I used on the cabinet. Tube Access Door Latch - The tube access door latch really only required a suitable metal thumbscrew, a washer, two nuts and a metal "arm" that's mounted to the thumbscrew shaft. I modified a spring contact arm I found in the junk box to work as the door latch arm. Very simple but it works fine. |
Mar 17, 2022 - EBY Binding Posts
- Six of these were required. I found one vintage original that
was unused and didn't have any engraving on the thumbnut. I had
a dozen or so more that I found digging through the junk boxes
but all of them had the engraved thumbnuts. I just chuck-up the
threaded rod end into a hand-drill and then use an emery board
(works better on bakelite than a file) to remove the engraving
and retain the proper shape of the thumbnut head. Then the
thumbnut is polished with 0000 steel wool. Wenol's can be used
for a high polish. If done carefully the end-result looks
somewhat original,...maybe a bit worn but still acceptable.
Reassembly Completed - Mar 20, 2022 - The shock feet were mounted and the license tag installed onto the cabinet. The tubes were finally tested and to my surprise the RF amplifier tube was a 6J7 instead of a 6K7. Although these tubes are almost identical, the 6K7 is a remote cut-off pentode and the 6J7 is a sharp cut-off pentode (different grid structures for AVC control or Manual control.) While these two tubes behave almost the same in this receiver the correct tube, the 6K7, allowed the VOLUME (RF gain) control to function better. The other tubes were correct and tested good. The receiver chassis was then installed into the cabinet (luckily, all 15 RH 6-32 machine screws were in the fruit juice cans.) |
| Disappointing
Results when
Testing with the Pixel Loop Requires Further Investigation
- To my surprise the Pixel Loop didn't work well at all with the
AR-8503. As with any loop, the signal levels available will be
less than a large outdoor wire antenna. But, the AR-8503
response level seemed to be very low - I couldn't even hear any
AM-BC signals. I thought it might be an impedance mismatch but
that didn't seem likely.
To confirm that the AR-8503 was functioning differently than the RAZ-1, I needed to operate the CRM-46092 receiver without the CRM-50092 preselector. Before doing that though, I checked the schematic for both receivers to confirm that they were indeed identical component for component (they are, or they should be.) The Pixel Loop operating with the CRM-46092 was quite different. Lots of strong signal levels with KPLY 630 "blasting" in requiring the VOLUME to be reduced to about 10%. The available sensitivity allowed tuning in the NDOT road conditions station on 530kc. Since it was daytime I was limited to the AM-BC for testing. The Pixel Loop will only go down to 150kc and actually begins to drop off response beginning at about 300kc. I went back to the AR-8503 just to verify the different performance and the only signal that could be heard was KPLY 630kc and that was with the VOLUME at about 75% advanced. |
AR-8503 Problems
- First, I verified that all of the ERIE resistors were the
correct value, or at least close. ERIE resistors can drift in
value depending on the circuit. All resistors were close in
value.
Schematic Error -
Checking further I discovered that the AR-8503 schematic that's
in the Sterling Radio Manual had the VOLUME pot and the
REGENERATION pot values at 750KΩ each. The actual values are
75KΩ each and those values are shown correctly in the RAZ-1
manual schematic and parts list. Both pots checked at about 75KΩ
in the AR-8503. Solder Problem
- Luckily, in checking the pot values I found that the buss wire
connection to the ground terminal on the VOLUME (RF gain) pot
had a broken solder joint. This was repaired. Further checks
didn't reveal any other problems.
Switch Contacts, Tuning Condenser Contacts and Spark Gaps
- I used DeOxit and a small paint brush to clean the band
switches and the grounding contacts on the tuning condenser. I
also hadn't installed the antenna spark gap so those two pieces
were mounted on the ANT and GND terminals. A quick test (out of the cabinet) using the Pixel Loop and KPLY 630 was tuned in with a very strong signal requiring the VOLUME control to be reduced down to about 10%. Several other AM-BC stations were tuned in from 650kc down to 530kc. Any further differences between the AR-8503 and the CRM-46092 can probably be attributed to the differences in the tubes. |
|
Radiomarine Corp.
- Model AR-8503 Radio Receiver SN: 42590 |
Nighttime Check
- Mar 27, 2022 - The AR-8503 was still out of the
cabinet but still connected to the power supply and to the Pixel
Loop. I was tuned to about 400kc and MOG 404kc was coming in and
with some adjusting of the controls the reception improved
significantly. Further tuning found GC 380kc in Gillette, WY,
RPX 362kc in Roundup, MT and RYN 338kc in Tucson, AZ. I only
listened for about five minutes since it was 2215hrs and this
was just a test to see that the reception using the AR-8503 and
the Pixel Loop had improved dramatically. It's also the end of
March and the conditions are slowly deteriorating as we get
further into Spring. Some lightning crashes were heard and
that's a sign that LW Season is coming to an end. AR-8503 Wrap-up - In looking at the photo to the left showing the completed AR-8503, note the Halli knobs and how at a casual glance or from a distance they look fairly close to the RMCA knobs that should be installed. The shock mounts are also fairly close but not exactly like the originals. This was a "refurbishment" that resulted in the AR-8503 looking good and performing quite well. The receiver is certainly a nice enough example that IF seven RMCA knobs became available I'd replace the Halli knobs "in a minute" and same goes for the shock mounts. The AR-8503 does receive all LW signals very well when using an antenna that is similar to what was used onboard the ship. That would be about a 200ft end-fed wire. That type of antenna will produce strong signals but also will produce quite a bit of noise. The shielded magnetic loop will not produce strong signals like a large outdoor wire antenna can, but the greatly reduced noise allows the listener to hear much weaker signals - a better signal to noise ratio. If possible, that is, if you're in a RFI-quiet area, use a large outdoor wire and the AR-8503 will respond well. But, in RFI-noisy areas, a shielded magnetic loop will allow receiving some LW signals. |
|
Radiomarine Corporation of America
NAVY DEPARTMENT - BUREAU OF SHIPS Serial Number: 65 - 1941 Contract (USN Accpt'd 1943) Includes: AR-8503-P Preselector SN: 65 |
|
| The USN RAZ-1 is the Navy version of RMCA's AR-8503. It is essentially identical to the commercial AR-8503 but was provided with the AR-8503-P Preselector and the RM-6 AC power supply. The RAZ-1 set profiled is SN:65. I really didn't have to do a lot of restoration when I first got the receiver in 2006. There were a few parts missing and the PS filters needed to be replaced was about all. Later, in 2022, I went through SN:65 completely and repaired a couple of problems that had plagued operation for quite some time. SN:65 now is fully functional and a pleasure to operate. |
|
The AR-8503 Regenerative Receiver was introduced around 1937 and was designed primarily for commercial shipboard installations. A matching pre-selector was also available later in production (dwg. dated 9-12-41) and the designation was AR-8503-P. Additionally, a Rectifier Power Unit (AC power supply) was offered, the RM-6, again later in production (dwg. dated Nov. 14, 1939.) Since DC operation was common onboard commercial ships, the AR-8503 was designed to be easily operated from a battery pack but Navy ships generally operated on AC and their preferred method of operation used the RM-6 to supply the required 6 volts for tube heaters, +22 vdc for the tube screen voltages and the RF amplifier variable cathode bias along with +90 vdc for the RF, detector (100K plate load resistor) and amplifier plates. Since many commercial ships of the time operated all of the radio equipment on DC voltages, the AR-8503 would be powered by the ship's battery-generator DC source for the tube heaters and dry cell batteries for the B+ requirements. Though I've never seen this on any AR-8503, supposedly the earliest versions had an emergency crystal detector that was mounted to the back of the tube access door and it could be connected into the receiver if ship's power was lost. This crystal detector option wasn't on the Navy RAZ-1 versions. The RMCA AR-8503 was available until about mid-1943 when its replacement LW receiver was introduced, the AR-8510. As far as the USN RAZ-1, it's possible that it was supplied to the Navy a bit longer since the AR-8503 was primarily used on either commercial ships or on Liberty ships during WWII. Sometime around 1941, the Navy Department wanted to install the AR-8503 on some of their smaller ships and a contract was issued for a small number of receivers. "RAZ-1" designated a complete set of equipment that included the CRM-46092 Receiver (AR-8503) with the matching CRM-50092 Pre-selector (AR-8503-P) and the CRM-20096 Rectifier Power Unit (RM-6.) The contract was NXs-94949 and the date was just five days before the attack on Pearl Harbor, Dec 2, 1941. As to the quantity of RAZ-1 sets produced, the manual-schematic
indicates that serial numbers exceeded 3615 with dwg. revisions dated as late as
July 15, 1943. The contract number was changed to NOs-94949 but the issue-date
remained the same (Dec. 2, 1941) and the later manual is dated July 1942.
Individual pieces of the RAZ-1 sets do show up occasionally but most often it's
just the CRM-46092 receiver that is found. Complete RAZ-1 sets (receiver,
pre-selector and rectifier power unit - all with matching serial numbers) are
very rare. The RAZ-1 set shown above is SN:65, an early production set that
interestingly wasn't accepted by the USN until May 12, 1943. Shown in the three
photos below are the data plates from the RAZ-1 equipment. |
|
|
|
|
|
>>> The CRM-46092 receiver has four tuning ranges covering 15 KC up to 600 KC. There are three bandswitches - two on the receiver and one on the pre-selector - that have to be utilized for changing tuning ranges. The National Type-N dials are scaled 0 to 100 and have a 180 degree layout. A tuning chart is provided in the manual to correlate the dial reading to tuned frequency. Coupling, Regeneration and Volume (RF Gain) controls are on the receiver's front panel and the pre-selector also has an RF Gain control. Audio output is provided for a single audio stage or for full audio output via two telephone jacks on the front panel. Output is designed for the Western Electric 509W earphones and, although any Hi-Z 'phones will work, the 509W phones seem to give the best immunity to noise. Baldwin Type-C "Navy" 'phones, while very sensitive, do reproduce a lot of bass (and power supply hum level) that makes copy of weak signals difficult. The +90vdc B+ will be on the 'phones when plugged into jack #2 since the phones are in series with the B+ to the 6F6 plate. The receiver case is shock mounted and is made of copper-plated steel that is painted gray wrinkle finish. The pre-selector case is made of aluminum and painted to match the receiver although it is not shock mounted. The power supply is a standard steel box painted gray wrinkle finish. The front panels of the receiver and the pre-selector are sheet brass that have been machine textured and then matte chromium plated. The receiver chassis is cad-plated steel. |
|
|
|
Left photo: The CRM-46092 chassis showing the large bee's wax dipped coils and the sparse layout of components. Four other coils (Band 3 RF Tuned Plate and Tickler and Band 4 RF Tuned Plate and Tickler) are located under the chassis. The tuning condenser is inside the shielded box in the center of the chassis. 2006 photo
|
|
| Performance - The
RAZ-1 can be very sensitive and almost any station on LW can be
tuned in, however the very best performance (for me, anyway) was
using a homemade remotely-tuned loop antenna. I've logged a lot
of NDBs with this RAZ-1 using the homemade four foot square,
remotely-tuned loop antenna. Of course, most of this RAZ-1 and
loop listening was well-over fifteen years ago when the "NDB
landscape" was heavily populated with lots of signals - quite
different from today's 2025 sparse (except for RFI harmonic
radiation,) desert-like signal landscape where more and more NDBs are decommissioned every month. The lack of any kind of
noise limiter or output limiter is sometimes a problem if local
noise is strong but loop antennas will greatly reduce the
noise (both natural and manmade.) To further reduce noise to an
absolute minimum, the Coupling should be set very close to zero
since that will help reduce bandwidth. The Regeneration should
be right on the oscillation point (autodyne detection) as this
provides a heterodyne for NDB detection and also will result in
the best selectivity. Then the received signals are peaked by
tuning the the Pre-selector to the received frequency. The Pre-selector won't appreciably amplify the incoming RF signal but it improves selectivity. The Trimmer control matches the RF output to the tuned detector input for the strongest signal response. The Pre-selector gain is usually set to about 85%, the Volume is set to about 25% normally but can be advanced as needed. These settings usually result in the best response of signal to noise along with the greatest selectivity. I've probably logged more NDBs with the RAZ-1 than any other LW receiver. However, that might be because it was one of the first LW receivers that I used when I started logging NDB stations in 2006. But, it could always be relied upon to pickup whatever was out there as long as reasonable conditions were present,...that would be, using my remotely-tuned loop antenna, earphones for reproducers and listening late at night during the LW Season. By 2018, I was using a Pixel Loop, a commercially-made shielded magnetic loop (SML.) SMLs will greatly reduce RFI noise and doesn't require constant tuning like the Remotely Tuned Loop did. There is a brief reception log for the RAZ-1 dated March 25, 1922 using the Pixel Loop antenna. It's below in the "More RAZ-1 Information." |
|
|
Finding the RAZ-1 - I first saw RAZ-1 sn: 65 in 1997 at the Galena, Nevada home of W3ON, John Ridgway. It was setting next to the SX-28 John was going to sell me (if I could lift it off of the table,...which I did.) I asked John if he wanted to also sell the RAZ-1, to which he replied, "You wouldn't take a longwave receiver away from an old Navy radioman, would you?" John was 85 years old and in the process of was moving back to Maryland. Over the next month, John sold me a National SW-3 receiver and a most of his test equipment but he held onto the RAZ-1. We casually and occasionally remained in contact after his move. John lived to the age of 93, becoming an SK in January 2006. To my surprise, in the summer of 2006, I got a 'phone call from a Maryland estate agent who said that they had found a letter among John's papers that stated that he wanted his radios and radio parts to be sent to the "Radio Museum in Virginia City, Nevada." The agent was calling me to see if I really wanted any of "this junk." I told them I did. The estate paid to ship the parts and equipment back out west. The shipping of the 22 boxes was spaced out over about a six week period. In the 21st box was the RAZ-1. Shipping had caused one small problem, one of the largest coils had broken from its mount. The large buss wiring had kept it in place and all that was required was to glue the mount back together and screw the coil form back in place. I acquired the correct shock mounts from N7ID (SK.) I did have to replace the filter capacitors in the rectifier power unit for quiet reception. Detailed Circuit Description: CRM-50092 Pre-Selector - The antenna input has a spark gap, a 100KΩ shunt to ground and is C-coupled using a 10pf capacitor to the input coils. There are four coils that are band switch-selected (BS) to connect as a parallel LC that is tuned with the tuning condenser and connected to the 6SG7 grid. RF Gain is controlled by varying the cathode to -AB (chassis) resistance utilizing a 250Ω in series with the 10KΩ variable R. +90vdc is connected to the screen and through a 10KΩ resistor to the plate. The output is coupled with a series LC filter to the antenna output terminal. Interesting Note: The engraved "AR-8503-P Radiomarine Corp" information on the front panel of this unit has "65" stamped in the serial number window. CRM-20096 - Rectifier Power Unit - The power unit uses a 5W4 as a rectifier tube. The power transformer secondary windings supply 6.3vac 3A, 5vac 3A and a HV winding for the +90vdc and +22vdc. The filtering is accomplished using a Pi-filter followed by a RC filter. Three 24mfd electrolytic capacitors are used and the RC resistor value is 1500Ω at 5W. The +22vdc is derived from the +90vdc using a series divider with a 5000Ω 5W and a 1500Ω 5W resistors. CRM-46092 - Radio Receiver - The antenna input has a spark gap, a 100KΩ shunt to ground and has an air variable C (Coupling) in series with the Antenna coils. The four BS-selected Antenna coils are parallel LC tuned using the Tuning condenser. The Antenna coils EM couple to the BS-selected RF grid input coils and are parallel LC tuned using the Tuning condenser. The Trimmer is an air variable to ground on the RF grid for exactly matching the Antenna coil LC to the RF grid coil LC. The RF amplifier tube gain is controlled by varying the cathode bias voltage (+22vdc) using a 750KΩ potentiometer and a 1500Ω cathode resistor. RF amplifier plate is routed through BS-selected tuned LC plate coils that are tuned using the Tuning condenser. The RF output is C-coupled to the grid of the Detector tube. The Regeneration control uses the +22vdc to provide a variable Detector screen voltage using a 750KΩ potentiometer. Four BS-selected Tickler coils have the Detector plate voltage routed through them to provide detector regeneration the level being dependent on the screen voltage adjustment (Regeneration.) The Detector plate is C-coupled to the 1st AF Amplifier tube grid. The 1st AF Amplifier screen is +22vdc and the plate is +90vdc. The 1st AF plate is C-coupled to the #1 phone jack and to the AF Output tube grid. The AF Output screen is +90vdc and the plate is routed through #2 phone jack and then to the +90vdc. This allows the 'phones to be in series between the B+ and the AF Output tube plate when #2 phone jack is used and also why Hi-Z 'phones must be used. The "ON-OFF" switch connects and disconnects the +6, +22 and +90 terminals from the radio circuitry. |
|
More RAZ-1 Information RAZ-1 with Pixel Loop - March 2022 TESTING INFORMATION |
|
Mar 25, 2022 - 2215hrs to 2225hrs PST MOG 404kc Montegue, CA RAZ-1 set, Pixel Loop, TRIMM 'phones,
cndx good. |
RAZ-1 with the Pixel
Loop
- I never had operated the RAZ-1 with the Pixel Loop. I
had very good results with my homemade remotely-tuned
loop so I wanted to see if a shielded magnetic loop
would also perform well. If the Pre-selector RF Gain is
is set too high, it will affect how the receiver
regeneration functions. But, as expected, the receiver
Coupling, Volume (receiver RF Gain) and Trimmer also
interact somewhat. Although "touchy" to adjust, I was
able to get good results with the Pixel Loop and the
RAZ-1 set. The reception log to the left are the results
of a ten minute listening test. I set up the RAZ-1 in the afternoon and tried the Pixel Loop on AM BC. The results were what was to be expected in that KPLY 630kc (Reno,NV) was extremely strong. Tuning was relatively easy since the signal was so strong. A little fine tuning and I picked up a weak AM BC station probably around 560kc. I used a RF signal generator for a signal source in the NDB range setting up for 404kc (for MOG.) Although the sig gen was easy to tune in with the RAZ-1, when that signal was removed, MOG's carrier could be heard but not MOG's MCW ID (this is normal reception of MOG during the daytime.) The test at night showed that the Pixel Loop can be used with the RAZ-1 set because the CRM-50092 Pre-selector is providing ample signal gain ahead of the RF amplifier in the CRM-46092 Receiver. The RAZ-1 and Pixel Loop combination is a bit "touchy" and requires constant small adjustments to the COUPLING, the TRIMMER, the REGENERATION and the VOLUME on the receiver. But, once the Pre-selector RF Gain is set to about 70 and the VOLUME (receiver RF gain) is set to about 50%, then the adjustments become less frequent. Also, it's very helpful to use both hands and tune the Pre-selector AND the Receiver simultaneously. It was very difficult to adjust the Pre-selector's RF Gain correctly. Lots of scratchy, intense noise with each movement of the gain control. So, more problems,... |
| Other RAZ-1 Issues CRM-46092 Receiver, for many years now I've noticed that the +90vdc terminal was very loose. Since 15 panel screws have to be removed just to be able to slightly pull the receiver out of the cabinet to tighten up the nut, I put off doing it. Also, the RAZ-1 had been relegated to "display status" for quite a while now so there wasn't any urgency to tighten the loose nut. Well, since we're testing the RAZ-1 everything should be repaired, so 15 panel screws were removed, the receiver slid out of the cabinet a few inches, then all of the 5/16 hex nuts on all of the terminals were tightened and the receiver slid back into the cabinet and the panel screws reinstalled. Further testing has shown the the Regeneration pot is very worn. It might clean up but if not the value is an odd one,...75KΩ. (Found one in the "pot box" - replaced Mar 29, 2022 - Regeneration now easy to control.) CRM-50092 Pre-Selector - The Pre-selector RF Gain control had always been very "noisy" so I thought it probably just needed cleaning. It was a style of potentiometer with a tandem ON-OFF switch. These types of pots are easy to disassemble for thorough cleaning. However, when disassembled the actual problem became painfully apparent - the carbon track had severe gouges, like a sharp tool had been used to create this half-inch long "canyon" in the carbon track. No amount of cleaning was going to fix this problem. The schematic and parts list indicate a value of 10KΩ, but, finding a 10KΩ pot with tandem switch wouldn't be too easy. The closest I could find in the potentiometer junk boxes was a 25KΩ (linear taper) with tandem switch. I couldn't assemble a "parts" pot-switch to create a 10KΩ with switch because internally the pot has to have the tab that actuates the switch and a standard pot doesn't have that internal tab. Since the original pot still had continuity it could be measured for value and it was a surprising 18KΩ! It's certainly possible that the gouge could also have some fractures associated with it that might have increased the total R somewhat. But, since the potentiometer is connected as a variable R for adjusting the RF gain by varying the cathode R, the maximum R of the pot shouldn't be too much of a problem as it will just result in a somewhat reduced range of adjustment (25KΩ to 250Ω instead of 10KΩ to 250Ω with the 250Ω being maximum gain.) So, the 25KΩ pot should function fine as the Pre-selector RF Gain control (a variable cathode resistance.) No issues with the pot installation. Both the receiver and pre-selector were reassembled, connected together and connected to the AC power supply. I pre-tuned to the top of Band 4 and as the RAZ-1 warmed-up, I could hear KPLY 630kc. An incredible difference! Now I could adjust the Pre-selector RF Gain just where I wanted it without the intense "scratchy-crackle" - just silently increasing the RF Gain. Of course, this also made the adjustment of the receiver much easier since the Pre-selector RF Gain could be exactly what was needed rather than a just a few places on the pot that happened to work. |
| The Final Test - Comparing
the AR-8503 to the CRM-46092 - Mar 28,
2022 - Rather than doing all of these reception
tests that attempt to compare performance differences between
the AR-8503 and the RAZ-1 by tuning in NDBs on different nights
with all of the variables that could be experienced with that
type of testing, I decided on a test that would be consistent as
far as the signal level. I decided to use the HP-606B RF Signal
Generator to output a measured signal that would be the same for
both receivers. The set up used the CRM-50092 Pre-selector with
both receivers. The same Pixel Loop for both receivers and the
same CRM-20096 power supply for both receivers. The only change
was using either the AR-8503 receiver or the CRM-46092 receiver.
The HP-606B output was connected to a ten foot long wire and the
Pixel Loop was located in another room upstairs with a distance
of about 35 feet between the Pixel Loop and the HP-606B
"antenna." This testing can be performed during the day, in
fact, it's better if it's performed during the day (no more
waiting until 2200hrs for testing.)
"Signal Level Reception Test" - The AR-8503 was tested first. The HP-606B was set to 388kc. This frequency was chosen because it's in the NDB portion of the MW band and it happened to be a frequency that was clear of any daytime switching or harmonic noises. The test was to first tune in the signal with the HP-606B output at about .030mv rms to the ten foot long wire. Once the signal was found, the receiver and pre-selector were further fine tuned and adjusted for maximum sensitivity. Then the HP-606B output was reduced until the signal couldn't be heard in the receiver 'phones. AR-8503 - The
received signal using the Pixel Loop was easily heard with the
output level at .003mv rms. Also, easy to hear at .002mv rms. CRM-46092-RAZ-1 - Interestingly, the CRM-46092 with the Pixel Loop performed exactly the same with the same results of "barely heard at .001mv rms" and disappeared below .001mv rms. NOTE: This isn't a "sensitivity measurement" of the receivers. What it does is to provide a known signal level with a specific radiator that's a specific distance away from the receiving antenna. The test setup is consistent and repeatable and that will help to evaluate the reception possibilities of a particular receiver using the Pixel Loop for the antenna. This type of test doesn't depend on receiving conditions that change from night to night. Sometimes our perception of a "really great performer" might be due to great conditions rather than the receiver itself. Additionally, testing like this can be performed at any time during the year and that's especially advantageous during the summer months when LW reception is very poor. |
|
|
|
Radiomarine Corporation of America AR-8506-B - WWII LF, MW and SW Superheterodyne Receiver The RMCA AR-8506-B was introduced during WWII with schematics dated November, 1942 and with the FCC approval for shipboard use dating from February, 1943. The AR-8506-B is a five band receiver capable of reception of LF signals from 85kc up to 550kc and medium/shortwave signals from 1.9mc up to 25mc. The circuit is superheterodyne and uses 10 tubes along with a NE-32 (G-10) neon lamp for voltage regulation (HFO.) The IF is 1700kc in order to allow the receiver to cover the entire 400kc range without interruption. Much of the ship's communications were in the frequency range of 400kc to 500kc and a standard IF of 455kc would have a gap in frequency coverage from about 430kc up to 475kc due to the IF operating at 455kc. Usually, shipboard superheterodynes will have IFs that are in the AM BC band area since this region of the spectrum wasn't normally tuned by the ship's communication receiver. The receiver can be powered by 115vdc or 115vac and can also be powered at 230v ac or dc using an external resistor unit, the RM-9. Tuning uses a 30 to 1 reduction vernier drive (counter-weighted) and there is an additional "band spread" function using a separate control. A built-in loudspeaker is front panel mounted and can be switched off by the operator if necessary. >>> |
|
|
>>> These WWII versions of the AR-8506-B don't have a front
panel ON-OFF switch because the receiver was powered up within the
communications console. The FCC approval for shipboard use indicated
that the AR-8506-B's HFO leakage to the antenna was <400pW and thus
would not interfere with other shipboard radio equipment and would not
radiate a signal of sufficient strength for enemy DF or detection. Note
in the photograph to the right the extensive shielding for the
receiver's front end to reduce HFO radiation. The AR-8506-B has an internal 1700kc wavetrap. The wiring and adjustment of the wave-trap should be checked if BC signal leakage is encountered. The wave trap should be adjusted on Band 3 for minimum response with a 1700kc RF signal input to A1 on the antenna input of the receiver. >>> |
|
| >>> If the wave-trap is correctly adjusted
and still there is BC signal leakage then using an antenna that is
"tuned" for the specific frequency desired should be tried. This could
be a resonant antenna cut for the specific frequency desired or an
antenna with an antenna tuner. The "tuned" antenna will be selective and
should reduce the BC interference. Like a lot of RCA receivers, the AR-8506-B doesn't have a standby switch (either remote or panel.) Most shipboard radio operations are receive on one frequency and transmit on another frequency using separate antennas so a standby wasn't really necessary. To use as a ham station receiver operating on the same frequency as the transmitter requires either an antenna relay with good isolation for the receiver in "transmit" or you can also use an electronic TR switch. Separate T-R antennas are also a possibility. In all cases, you'll have to reduce the RF Gain and AF Gain or switch off the loud speaker if you're on Voice transmission to avoid feedback. These receivers were integrated within a shipboard communications console, generally the 4U, that contained two transmitters, another receiver capable of VLF reception (AR-8510,) an emergency receiver (crystal detector receiver,) a power control switching system that allowed battery operation or ship's power operation, motor-generator operation, various alarms and other equipment necessary for radio communication at sea. Most 4U consoles were installed on Victory ships and other merchant ships during WWII. Shown to the right is the RMCA 4U marine radio unit. Note that the receiver to the left of the typewriter is the LF-MW-SW Superhet receiver, the AR-8506-B, and the receiver to the right of the typewriter is the VLF-LF-MW Regenerative TRF receiver, the AR-8510. The two transmitters are mounted directly above the receivers. Note the RCA Radiogram paper in the Royal mill-typewriter. Photo is from Sterling's "The Radio Manual" Fourth Edition. |
|
|
|
|
Radiomarine Corporation of America Model AR-8510 - WWII VLF, LF and MW Regenerative TRF Receiver |
| The AR-8510 was
the VLF, LF and MW receiver that was usually installed into the various
types of Liberty ship radio communication consoles. The receiver could
be powered on ships that
provided 115vdc or 230vdc. If 115vac was to be used then the RM-23
Rectifier Power Unit (power supply) was used. There was also an RM-37A
Receiver B+ Supply Unit that provided 90vdc output from the ship's
115vdc power. This was to be used if it was necessary to conserve the B
batteries that normally provided the +90vdc for the B+. The AR-8510
requires 6.3 volts at 1.8A (AC or DC) and 90vdc at 15mA. The vacuum
tubes needed are four 6SK7 tubes and one 6V6G or GT. Unlike the
AR-8506-B, the AR-8510 does have a front panel ON-OFF switch to allow it
to be "switched off" while the AR-8506-B would remain powered up through
the communications console. Circuit Description - The AR-8510 is a five tube regenerative receiver that tunes from 15kc up to 650kc in four tuning ranges. Two TRF amplifiers are used with a Regenerative Detector and two stages of audio amplification. The RF amplifiers use a combination of tuned grid input and tuned plate output using a three-section ganged tuning capacitor. The antenna switch allows the user to select which receiver will be connected to the antenna - either the AR-8510 or an emergency receiver (usually a small Crystal Set that can operate without ship's power.) The audio output can drive the panel mounted loud speaker or headsets either simultaneously or, using the Loudspeaker switch, the panel speaker can be turned off. The receiver requires a separate power source of which many types were available. Various types of battery combinations could be utilized with either the RM-2 or the RM-4 Battery Control panels. These functioned on ships that provided 115vdc or 230vdc power. If 115vac was to be used then the RM-23 Rectifier (power supply) was used. There was also an RM-37A unit that provided 90vdc B+ output with a 115vdc input from the ship's power. This was to be used if it was necessary to conserve the B batteries that normally provided the +90vdc. The AR-8510 requires 6.3 volts at 1.8A (AC or DC) and 90vdc at 15mA. The vacuum tubes used are four 6SK7 tubes and one 6V6G or GT. The AR-8510 was approved by the FCC for shipboard use in 1942. The schematic drawings are dated 1943. The AR-8510 shown in the photo above is dated NOV 1944 and has a serial number of 2774. The AR-8510 serial doesn't follow the typical RMCA format of having the manufactured year as the first two digits of the serial number. Instead the month and year are stamped on the data plate. This particular receiver was installed in an RMCA communications console so it doesn't have a cabinet. However it does have a bottom cover that incorporates embossed dimples for feet that allow the receiver to set on a table without scratching the table top. |
|
|
|
The AR-8510 could be provided with a cabinet and shock mounts if
it was to be used as a "stand alone" receiver. However, if it was going
to be installed into a shipboard communications console (as most were)
then the cabinet and shock mounts were not provided. Many AR-8510
receivers were part of the shipboard 3U transmitter console that
included a 200W transmitter, an emergency crystal receiver, battery
charger switching, clocks and more. The 4U console used
the RMCA AR-8506 MW-SW receiver with a larger 500W transmitter. The 5U
console had both receivers installed along with transmitters and
auxiliary equipment. Mackay Radio and Telegraph Company also supplied
Marine Radio Consoles MRU-19 or 20 that had their equipment installed.
Most AR-8510 receivers were console installations during WWII onboard Liberty ships. Post-WWII installations were generally on commercial ships. The AR-8510 found a lot of use and longevity with production unbelievably still going on in 1960. Some receivers were still in use onboard some old oil tankers as late as the 1990s. Most frugal ship owners wouldn't replace the radio gear as long as it still functioned or could be cheaply repaired. Unfortunately, most AR-8510 receivers led a pretty hard life and the sea environment didn't help preservation. Most examples have been worked on or have missing parts (or non-original parts.) The AR-8510 shown in the header photo is cosmetically restored with nearly all original parts. The exception is one capacitor under the chassis, the speaker grille and the RCA pointer knobs. The paint job on the front panel is VHT Gray wrinkle finish which is slightly darker than the original RMCA gray. I was given the AR-8510 shown in the photos as payment for some radio repair work. It probably was taken off of one of the Liberty ships that were part of the "moth-balled" fleet that was moored outside of Benicia, California since the receiver originally was obtained from the SF Bay Area. The "as found" condition was fairly good considering how the ships were taken care of (they weren't.) Of course, the front panel has been repainted in the past - probably with a brush. The perf-metal grille had more than its share of paint applied (and it wasn't original either.) The receiver came without any type of power supply (the RAZ-1 rectifier power unit RM-6 can be used as a power source.) |
| This AR-8510 required a little bit of work to get it
operational. Bands 1 and 2 functioned okay but needed alignment.
Band 3 and 4 were non-functional due to broken leads from the coils
that are in the plate circuit. The open coils resulted in an
absence of plate voltage to the first RF amplifier when bands 3 or 4
were selected. I had to remove the coils and rebuild them then finishing them
off with a re-waxing job. After reinstalling, bands 3 and 4 had to
be aligned. Cosmetic Restoration (2014) - I was impressed with the AR-8510's performance but its appearance, while not bad for a vintage "as found" condition receiver, could certainly be improved upon. The photo to the right shows how the AR-8510 looked when received. Close inspection revealed that the "brush" paint job had covered up all of the nickel plated hardware. All screws heads, phone jack nuts, pilot lamp mount, even the non-original aluminum perf-metal grille were all painted over. My plan was to dismount the front panel and give it a new gray wrinkle finish paint job. Luckily, VHT had just started to offer Gray Wrinkle again but it wasn't locally available. I had to buy the VHT paint off of eBay from one of the large auto supply houses in Ohio (JEGS.) Removing the panel was involved but not difficult. Some of the vernier mechanism's hardware is unique - the three stand-offs are relieved to allow clearance for the vernier housing. Any difficulty in disassembly was due to corrosion and this was only in the knob set screws. A combination of heat applied with a solder iron tip and WD-40 worked to loosen all but one set screw. That particular knob had to be broken (it was a non-original standard pointer knob) in order to remove the set screw. With the panel devoid of parts it could then be dismounted for stripping and repaint. |
|
| Painting Gray
Wrinkle Finish - I used a methylene-chloride based stripper
(JASCO, but I don't use this type of stripper anymore) to remove all paint. As the slopped-on dark gray came off I
could see why the panel had been repainted. The original WF had several
areas where paint had come off leaving bare metal. I've seen this before
on a few vintage WF paint jobs. Since the WF is applied to bare metal
(you don't use primer or it will spoil the WF pattern) the bare metal
surface has to be in good condition and extremely clean. Sometimes this
didn't happen with the original paint job and then, usually years later,
the WF paint begins to flake off in large pieces. This must have been
what happened to this AR-8510.
After stripping, I usually give the metal a good scrub with lacquer thinner using a 0000 steel wool pad. Then finish with another wash with lacquer thinner and paper towels. This usually gets the panel ultra-clean and the surface conditioned for the best results using WF paint. Since both sides of the panel were originally painted with WF, I had to think of how I was going to paint the second side without ruining the first. I could hang the panel and paint but this would make heating a problem. Using the threaded bosses that are for the panel mounting thumb screws, I threaded in long screws that would keep the panel surface raised so the paint job wouldn't be compromised when the back of the panel was painted. >>> |
>>> Wrinkle finish must be heated to activate the wrinkling. I use two clamp-lamps with the aluminum bell reflectors with 100W incandescent lamps. These are placed approximately 10" above the paint job to apply an even low heat. In about 5 to 10 minutes the wrinkling will begin. I standby with a hand-held heatgun to apply more heat as needed. Usually the lamps won't get enough heat to the corners of the panel and this is where the heatgun comes in. You can start to wave the heatgun as soon as the lamps are turned on. Don't hold it on one section but move the gun around just applying a little additional heat. You only want a gentle heating. Too much heat will "gloss" the wrinkle and spoil the paint job so be sure to not hold the heatgun on one spot but keep it moving. As the wrinkle progresses you can heat the areas that seem to be slower in wrinkling. This is usually the edges or the corners. Just a little additional heat will get the wrinkle process going in these "tough to heat" areas. After the paint has completed wrinkling, remove the heat lamps and let the panel cool down. After 30 minutes you can gently handle the panel but be careful. Now the back side is painted using the same process. It's difficult to do both sides at the same time because of the surface heating that is required. Though you will get heat transfer by conduction to a back side you won't really be able to see what's going on and won't be able to control the wrinkle. You also can't apply the WF paint to the panel when it's still hot, so painting the backside immediately is not recommended - let it cool down. If the WF is painted on a hot surface it will begin to wrinkle right away and you'll end up with an uneven pattern. Slow, controlled heat is necessary for a professional looking WF paint job. When the WF painting is complete you still have to be careful. The WF paint is still very soft. Let the paint set up overnight at least - a couple of days is better. You can go ahead with remounting after a day or two as the WF will be fairly hard by then. The WF will reach its maximum hardness after about 30 days. |
|
 |
| Grille Replacement
- The installed perf-metal grille was not original and didn't even
resemble the "backwards swastika" pattern that was original to the AR-8510.
Finding an original pattern perf-metal grille was going to be impossible
but I was pretty sure I'd seen a similar pattern somewhere. I noticed
that the Hammarlund Super Pro and HQ-120 speaker boxes use the same
"backwards swastika" pattern but these speakers were not something I wanted to
"junk" just for the grille. With a little searching I found a junk Bud
metal box for a speaker that had a grille with small rectangular
cut-outs that were similar to the original "backwards swastika" pattern. This
grille was cut to the correct size and was painted light gray smooth
finish to contrast with the panel's WF. For the grille cloth I robbed
the black loose weave cloth that's found in old (junk) Zenith Transoceanic
radios. Grille Board - In order to have the grille cloth have proper support a cardboard grille backing is necessary. I made one out of heavy cardboard that was very similar to the backboard on a paper tablet. Actually, I used cardboard packing dividers but the material is almost exactly the same as that found on tablets. Once the cardboard is cut then the cloth can be mounted. Use either spray adhesive or a light coating of Elmer's Glue just on the edges. Pull the cloth tight and let the glue set up. Use a soldering iron hot tip to burn through the grille cloth at the mounting holes. See photo right for how the grille board and grille cloth look. |
 |
|
|
Hardware -
Almost all of the screw heads were painted over with the dark gray
paint. Originally these are nickel-plated brass screws so the paint
needed to be removed. This was easily done with the use of a brass
"toothbrush." Brass will not scratch the nickel and will remove the
paint without damaging the screw. All control nuts and washers were also
brass-brushed. The nuts on the phone jacks were replaced because of poor
condition. Tags, Knobs and Dial Cover - All tags were cleaned before remounting. The pointer knobs were replaced with RCA-type knobs. Although these aren't original either, they are at least an "RCA product." The bandswitch and tuning knobs are the correct "Radiomarine" knobs and that is what the other knobs should look like. The dial cover is heavy plastic and is held in place with six screws. You have to mount the tuning knob before you mount the dial cover since there isn't access to the knob set screws with the dial cover in place. You have to use a long screw driver because you will have to access the set screws behind the unmounted dial cover. Place the dial scale on the tuning shaft followed by the dial cover and then the tuning knob. Slide the knob up the tuning shaft making sure you don't have it so far you can't get the screw driver behind the dial cover. Notice that you can access the knob set screws behind the dial cover. Tighten the set screws and then mount the dial scale and the dial cover using the six screws. Be sure when mounting all plastic tags and the plastic dial cover to not over-tighten the screws as this will crack the plastic. Snug the screws and then very slightly back the screw out to relieve the pressure. This way the plastic parts won't move easily but will have room for expansion and thus won't crack. |
| Reassembly and
Testing - I let the front panel WF paint set up for three
days before doing the reassembly. Even then you have to be careful
because the paint is still not at full hardness. With the AR-8510 fully
reassembled I was ready to test. First, I listened at 11:00PM local time using the 135' CF Inv Vee with the 98' feed line shorted as the antenna. I tuned in several NDBs from the Western US. I also copied NY and XX from BC, Canada. Lightning crashes were coming about every two seconds so copy was difficult. August conditions are very poor for MW signals but the AR-8510 still managed to receive several NDBs. The next morning at 6AM, I tuned in JJY on 40kc coming in fairly strong. On WWVB 60kc or JJY 40kc, you have to greatly reduce the RF Gain in order to hear the pulse encoded signal. Too much gain will allow the noise to overpower the signal. The same is true for NAA and NLK, the USN MSK RTTY stations on 24.0kc and 24.8kc. |
Since the AR-8510 was the replacement for the AR-8503, it's fair to compare the two receivers. First, with a direct readout dial there's no need for the charts and graphs that are necessary for finding where you're tuned on the AR-8503. The preselector is built-in with the AR-8510. Also, only a single band switch is necessary on the AR-8510 while two band switches are used on the AR-8503 plus a band switch on the preselector. Sensitivity on the AR-8510 is about the same as the AR-8503 with preselector. Regeneration action is very similar in that it's a very sharp adjustment between maximum sensitivity (either non-oscillating or oscillating) and any adjustment below either point greatly reduces sensitivity (this is typical of regenerative detectors though.) The AR-8510 seems to hold its adjustments better across the band especially the Trimmers that only require a slight adjustment from one band end to the other. This is expected since it's part of the alignment process. I find that the loudspeaker is actually pretty good for some reception. If you want to use Hi-Z phones, it's better if you leave the loudspeaker on. Without the speaker load the 'phones seem to respond to more noise than signal. All in all, the AR-8510 is a fine performer and should do an excellent job in the coming LF season. NOTE: I do believe that the metal grille looks awful in the light gray paint. I'm planning on a repaint with the grille matching the panel VHT Gray Wrinkle Finish. I'll post a new photo when complete. |
|
|
Radiomarine Corporation of America LS-1 Loud Speaker SN:4475
The probable year of manufacture is 1944. The switch selects either low impedance or high impedance. DCR in low Z is 5.0 ohms and DCR in high Z is 640 ohms. This would imply that the nominal impedance would be about 10 Z ohms for low and about 8000 Z ohms for high. The case is made of 3/8" solid oak with dovetail joints. This example has the wood painted black but originally the cabinet was painted gray. The speaker is a 3" diameter PM and a matching transformer is also inside the cabinet. I'm not sure whether the LS-1 was for testing applications or if it was for use as a temporary loudspeaker. The cable is about three feet long with a standard 1/4" phone plug on the end. The knob may not be the original style. It's probable that the LS-1 was primarily for testing purposes since all shipboard radio operators used headphones for reception and most RMCA receivers had a front panel loudspeaker that could be switched on when needed. |
|
|
The Post-WWII era |
|
|
|
Radiomarine Corporation of America AR-8506-B aka: R-203/SR and R-203A/SR - Post-WWII Version After WWII, the AR-8506-B continued to be offered by RMCA for maritime use on various types of ships. The post-war versions are somewhat different in appearance in that the individual celluloid control identification plates are replaced with a "raised letter" type of panel nomenclature. Additionally, the data plate was removed and the manufacturing information became part of the front panel nomenclature. Since many of the post-WWII AR-8506-B receivers were not going into communications consoles, a front panel ON-OFF switch was provided. The AR-8506-B shown in the photo above is from 1953 and shows how the later versions looked when installed in the table top cabinet (with shock mounts.) It's interesting that RMCA was still building and selling a ship's receiver design that was over ten years old. Ship owner's reluctance to replace radio gear was probably why RMCA could still find buyers for obviously dated designs. Since the ship owners didn't want to replace their old RMCA consoles anyway, as a consequence, the associated radio equipment was in-use well beyond the normal life-span with examples of the AR-8506-B and AR-8510 still in use as late as the 1980s. These later post-war, stand-alone (not in consoles) AR-8506-B receivers were also used at the RMCA coastal station KPH for various purposes. After WWII, the Signal Corps wanted their own version of the AR-8506-B. RMCA built the R-203/SR and the R-203A/SR for the Signal Corps. This version had an engraved front panel that was painted gray wrinkle finish. Standard RCA-type knobs were installed rather than RMCA knobs. The shock mounts were bolted to a large metal base plate. The optional Noise Suppressor F-91/U and the Fixed Resistor MX-1024/U were included with the receiver. The differences between the "non-A" and the "A" versions are the "A" had a separate ON-OFF switch while the "non-A" had the power switch as part of the RF gain control. Several resistor values differ between the "A" and "non-A." A different Noise Filter was supplied with the "non-A." The number of ground post differs between the two types of receivers. The Signal Corps manual is TM11-875. The AR-8506-B circuit used a TRF preselector stage using a 6SG7 metal octal tube. The Mixer stage was also a 6SG7. The HFO used a triode 6J5 metal tube. Three IF amplifier stages were employed using three 6SG7 metal octal tubes. The Detector/AVC and 1st AF amplifier stages used a metal 6SQ7 duplex diode-triode tube. The BFO used a triode 6J5 metal tube and the BFO adjustment was a round knob located on top of the BFO shield-can. The audio output tube was a glass 25L6GT tube (2 watts max or 750mW undistorted audio.) The rectifier was also a glass tube, a 25Z6GT. Additionally, a NE-32 neon bulb was used as a voltage regulator for the HFO. The light from the NE-32 doubles as the pilot lamp illumination. The IF operated a 1700kc to allow continuous coverage of the 400kc to 500kc part of the spectrum. The 1700kc IF will place the image at 3.4mc from the desired signal which helps reduce the image response at higher frequencies. The AR-8506-B has an internal 1700kc wavetrap. The wiring and adjustment of the wavetrap should be checked if BC signal leakage is encountered. There are six AM BC stations in the USA that operate on 1700kc located in Texas (2 stations,) Iowa, Alabama, Florida and New York (all are 10KW day and 1KW or less at night.) There's another 1700kc AM BC station located in Baja California, Mexico. Since 1700kc is so sparsely utilized, the chances of encountering 1700kc AM BC leakage is slight. However, things can always change so proper adjustment of the wave trap should be performed. The wave trap should be adjusted on Band 3 for minimum response with a 1700kc RF signal input to A1 on the antenna input of the receiver. If you're unlucky enough to be in an area with a strong 1700kc station and you have the wave trap adjusted correctly and there's still BC signal leakage then using an antenna that is "tuned" for the specific frequency desired should be tried. This could be a resonant antenna cut for the specific frequency desired or an antenna with an antenna tuner. The "tuned" antenna will be very selective and should reduce the BC interference. It's also possible to install an external wave trap between the antenna feed line and the receiver antenna input for severe interference. Since the AR-8506-B is a "transformerless" AC-DC circuit, always operate the receiver using a 1:1 isolation transformer when powered by the house AC line voltage. This will prevent AC line to ground conflicts through peripheral equipment grounding. It's also possible to build a combination isolation transformer and a rectifier-filter circuit to operate the AR-8506-B on +115 vdc (see RMCA AR-8516 section further down for details on the MM-555140-B combination isolation transformer and DC filter unit that provides +115vdc to operate that receiver.) I've used this AR-8506-B as a station receiver in combination with an ART-13 transmitter. Other than the set-up problems mentioned in the proceeding paragraph, the receiver has good performance capability. Since the ART-13 has an internal T-R relay that utilizes a vacuum switch for isolation, the standby function can be achieved by reduction of the RF gain and AF gain controls (if phones are being used.) If only the panel loudspeaker is being used, then the Loudspeaker switch can be placed in the OFF position when transmitting. I used an isolation transformer inline with the AC input to the receiver to assure that no AC power versus ground conflicts would occur. |
|
|
|
Radiomarine Corporation of America - Model AR-8510 - Post-WWII Version The AR-8510 was provided with a cabinet and shock mounts if it was to be used as a "stand alone" receiver. The AR-8510 shown in the photo above is the post-WWII version and the photo is from the manual. Like the AR-8506-B, the AR-8510 replaced the round celluloid nomenclature plates and went to an embossed nomenclature panel. Note the "raised letter" nomenclature which is not used on earlier versions. Ease of maintenance during WWII would have had the receivers fitted with the celluloid tags that could easily be removed for repainting the panel or replacement if damaged. After WWII, the "raised letter" panels were probably used since the receiver wouldn't have to endure the rigors of war-time use. Unbelievably, the production of the AR-8510 continued into the 1960s and actual use of the AR-8510 lasted quite a bit longer. It wasn't uncommon to find AR-8510 receivers still being used on old oil tankers as late as the 1990s. |
|
Radiomarine Corporation of America - Model AR-8711 Direction Finder - 1947 Post-WWII found that there was a growing interest in the operation of small private boats and in piloting those craft on lakes, bays and maybe even some distance off the coasts. Larger private craft that provided sleeping berths and storage for supplies might be used for over-night to several day excursions that could involve operating the craft long distances from shore and perhaps in the dark hours or maybe in fog-bound waterways. Many small businesses were started for charter trips, excursions and fishing trips that may have required a moderate size boat to travel long distances in unfamiliar waters. At any rate, Radiomarine Corporation decided that there was enough of a market to build and sell small, easy to operate, radio direction finders that could be used to find correct bearings for maritime navigation. Most of the larger bodies of water had marine beacons at various locations (many associated with lighthouses) and some buoys were equipped with radio beacons. At the time, Lighthouse Ships, equipped with radio beacons, were also in use as navigation aids. Navigation charts usually had the locations and radio identifications shown. It was also possible to use a known AM-BC station to provide a "beacon" to a city. So, the intention of the AR-8711 was to offer owner-operators of cabin cruiser type boats a method of navigation if the visual options were unavailable. The original selling price of the AR-8711 was $149.00 in 1947. Circuit - The AR-8711 is an eight-tube superheterodyne receiver that is designed to be used as a direction finder. Since the intended operation was to be onboard a boat, four different types of power supplies were available as options. There were three DC types of supplies for 6vdc, 12vdc or 32/115vdc operation and one AC supply for 115vac input. The DC supplies were vibrator based and the 32vdc unit could also use an adapter (RM-93) for 115vdc input. The AC supply used a 6X4 rectifier tube and the 32vdc DC supply used a 0Z4 rectifier tube. The 6vdc and 12vdc just used filters for the switched voltage out of the transformer. Power input to the AR-8711 was via a 21 pin Cinch-Jones male plug on the rear chassis. The same type of connector was used on the power supplies and an interconnecting cable was supplied. The loudspeaker is built into the front panel. The eye-tube is a tuning aid and requires the AVC be turned on for it to function. DFing - The AR-8711 is a radio receiver that operates as a direction finder. Two antennas have to be connected for an indication of "true" direction. The simple wire or vertical antenna is used as the "Sense" antenna and it is always in the circuit except when the BALANCE control is in the OFF position. The integral loop antenna provides a "figure-8" pattern that is bi-directional with signal level peaks off the ends (in line with the loop axis) and deep nulls in signal level off of both sides. The "loop only" is in the circuit when the BALANCE control is switched to the OFF position. The loop alone provides ambiguous directional information which can be used in some types of navigation. When the two antennas are combined within the receiver front-end, then a "cardioid" antenna pattern results, this occurs at any position of the BALANCE control except OFF. The advantage of the "cardioid" pattern is that only one null is produced which can be used for finding "true" direction. The "null" position is very narrow and much more accurate than the "peak" response, which in a "cardioid" pattern is extremely broad. NULL was used to increase the sensitivity of the receiver for a better null indication. DF gain was used for controlling IF gain when DFing since the AVC was disabled for DFing. Since the Loop and the Sense antennae are always connected the DF procedure is a little different than the standard DF receiver. When finding the null, the loop's "red band" - setup to be the "forward" part of the loop - is either pointing towards the signal source or 180º away from the signal source. The BALANCE is switched to SENSE and the null "fine tuned." Then the operator slightly rotated the loop to increase the compass reading. If the null decreased in strength then the Loop was pointed at the signal source. If the null strength increased then the Loop was pointed 180º away from the signal source and the opposite compass pointer indicated the true direction of the signal (two opposing pointers, one red and one white are part of the compass.) |
|
| Using the AR-8711 - If the boat pilot knew the approximate direction of the desired port, then bi-directional, that is, just the loop antenna could be used by switching BALANCE to OFF. By using a known AM-BC station that was located in the same city as the port as a "beacon" the boat pilot could tune in that AM-BC station on the AR-8711, then setting the loop athwartship, he would steer the boat for the minimum signal response from the AR-8711. This would have the boat heading directly for the AM-BC station's antenna. Listening to the AR-8711, if the boat pilot heard the signal increase in strength, he knew that he was slightly "off course" and could correct by steering the boat for minimum signal level. This method worked when there was an AM-BC station in the city that the desired port was located in and that the boat pilot knew the approximate direction of the city from the boat. |
 |
If a marine
beacon or lighthouse ship (they were usually equipped with a radio
beacon also) was in the direction of the desired course but the actual
"true" direction of the beacon wasn't known (you're lost, in
other words) then the "true" direction could be determined by using Sense DFing. This required the sense antenna in combination with the loop to
provide a cardioid pattern. Once
the beacon was tuned with BALANCE ON, then the signal was
tuned for absolute minimum, the null. The compass bearing is noted and
then BALANCE switched to SENSE and the null fine tuned. When "true
bearing" was found the loop axis was pointing at the signal source and
the bearing could be read on the compass. If you were really
lost, using two maritime beacons (their positions and frequency shown on
the navigation chart,) a bearing could be taken of each beacon and a
line drawn for each beacon's bearing on the chart and the point of
intersection would be the your location (a form of triangulation.)
All of this assumes that the DF radio was setup in the boat with the loop axis oriented for 0º being the bow of the boat and that the boat was also equipped with a magnetic compass that gave the direction of North plus or minus whatever the magnetic variation was in the area for "true" North. From the magnetic compass reading with compensation for magnetic variation and the DF bearing, a correct course could be plotted for the boat to be steered towards the beacon. When the AR-8711 was operated by an experienced navigator it could provide a lot of information. For example, distance to a coastal radio beacon (at a lighthouse for instance) could be determined by taking two radio DF bearings (some distance apart) as the boat passed the beacon. The bearing angles measured and the distance traveled between the first and second bearing could be used to calculate the distance to shore where the beacon was located, essentially letting the boat pilot know how far out from land he was (the distance traveled defines the base of a triangle and, with the two bearings providing the angles, the two remaining sides (distances) of the triangle can be calculated thus providing the distance to shore from either bearing point.) At distances greater than 50 miles, when using Mercantor charts, a slight correction factor had to be added or subtracted because of the distortion that is inherent in Mercantor projection charts. Most of the larger cabin cruiser boats that would have had a need for the AR-8711 would also have been equipped with other marine radio gear for ship to shore communications. |
|
|
|
Radiomarine Corporation of America Model AR-8516 LF-MW-SW Communications Receiver
RMCA's AR-8516 was introduced in 1958. When compared to its popular predecessor, the AR-8506-B, the AR-8516 was a quantum leap in the technology of radio receiver design and construction. Features included "almost to the kilocycle" dial accuracy, a dual-frequency variably-tuned IF plus single, double and triple IF conversions, use of a 3.1kc Collins Mechanical Filter, two AGC time constants, good SSB demodulation,...on and on. When fully rebuilt and in excellent mechanical condition with a fresh IF/RF alignment, the AR-8516 is a marvelous receiver providing superior performance in all modes of reception. But, it's a very complex receiver that had to endure the rigors of life onboard a ship at sea and be able to function reliably in that harsh environment. Nowadays, finding a good condition AR-8516 to start with is hard enough and then the difficulty of the electro-mechanical rebuild will require a substantial amount of technical experience to ultimately achieve what the receiver's design is capable of. The manual isn't very much help either with poor layout, typical errors, confusing alignment instructions and no small component location drawings. The following is my typical OCD-detailed write-up of the restoration process,... |
|
|
|
|
Circuit Description - The AR-8516 was a thoroughly modern (well, for the late-fifties,) complex communications receiver that used 18 tubes in a single, double or triple conversion superheterodyne circuit. The receiver used a variably-tuned IF covering 1.09 to 3.09mc or 2 to 4mc. The fixed-frequency IFs were at 455kc and 45kc. Single conversion was used in the first five bands covering 80kc up to 4mc. Bands 1-4 span different widths of the LF and MF spectrum and, along with Band 5, these single conversion bands tune from left to right being from high to low in frequency. From 2mc up to 30mc is covered in fourteen 2mc-wide bands. Band 5 (2-4mc) is single conversion but Bands 6-18 are double conversion. The double conversion bands (Bands 6-18) tune left to right being low to high in frequency. The total frequency coverage is from 80kc up to 30mc in 18 tuning ranges. Seven crystals are utilized in the Crystal Oscillator with only two crystals used uniquely. The other five crystals are used in various fundamentals and harmonics for a total of 13 crystal oscillator frequencies. The RF amplifier, Crystal Oscillator and Mixer 1 circuit function as the RF input for bands 6 to 18 with the output going to the Variable IF tuning that then provides a fixed 455kc output. The RF input section is bypassed when using bands 1 to 5 and the incoming signal goes to the Variable IF which combines with Mixer 2 and the selected VFO to provide a 455kc output. Two separate VFOs are used with one used for Bands 1, 2 and 3 and the second VFO used for Bands 4 to 18. Main tuning gear reduction is 41.7 to 1 and the SECTOR gear reduction is another 3.8 to 1. The resulting tuning is a constant "bandspread" type at almost a vernier tuning rate. The dial readout is "band-in-use" and employs a rotating dial drum that is ball chain-coupled to the band switch. The smaller dial to the right is the kilocycle-logging scale which for Bands 1-4 is a logging scale, but, since the Bands 5-18 are 2mc linear scales for each band, if the receiver is well-aligned, then the this dial will readout almost directly "to the kilocycle" for these bands. For example, if the slide-rule dial pointer is set in between 14.1 and 14.2 and the logging dial reads 67, then the tuned frequency is 14.167mc. The kilocycle dial index is not adjustable. The accuracy of alignment will determine just how usable the kilocycle dial ends up being. Spec from 4-30mc is "within 10kc." I was able to achieve an average tracking accuracy of between "dead on" up to about 5kc off. Below 4mc the accuracy spec is 0.5% which is easily attained during alignment (at 4mc the spec of 0.5% is 20kc.) Bandwidth is selectable from 6kc down to 100hz (five steps) in Bands 3 to 18 but in Bands 1 and 2 the bandwidth is selectable only from 1.5kc down to 100hz in three steps (6kc and 3.1kc, if selected, won't change from 1.5kc bandwidth.) The selectable bandwidths provided are 6kc, 3kc, 1.5kc, 800hz and 100hz. The 3.0kc bandwidth utilizes a Collins 3.1kc mechanical filter. The 45kc IF is utilized for two purposes. In Bands 1 and 2, 45kc is the IF. In Bands 3 to 18, the 45kc IF is utilized as part of the selectivity bandwidth operation and results in triple conversion in the IF section if 1.5kc, 800hz or 100hz bandwidths are selected. Two types of detectors are used with a standard vacuum tube diode detector used in Bands 3 to 18 and a crystal diode detector (heavy duty 1N34 germanium diode) utilized in Bands 1 and 2. Two BFOs are also used with the 455kc BFO being adjustable +/- 2kc but the BFO for 45kc (which is used just for bands 1 and 2) utilizes a fixed-frequency (adjusted for a 1kc heterodyne or 44kc.) When SSB is selected the 455kc BFO output is increased (compared to the output in CW) for better demodulation of suppressed-carrier signals. AGC has selectable fast and slow time constants. The audio output frequency response is 200hz to 3000hz (-4db down) and the audio power available is 250mW with 3.2Z or 600Z impedance outputs provided. A 500kc Calibration marker signal is available that functions from the 500kc Crystal Oscillator that also provides mixing with 455kc IF for the 45kc IF. A carrier level meter (showing db above 1uv) is provided. The ANTENNA (trimmer) control is connected to RF/Mixer 1 and only functions when the receiver is on Bands 6-18 (4-30mc.) Radiomarine also offered a very similar receiver that was designed for AC line power application at any land-based communications operation (like RMCA Coastal Stations.) The CRM-R6A has many slight differences but it's basically the same type of design as the AR-8516, both electronically and mechanically. Power Requirements - Being a commercial shipboard receiver, the AR-8516 is designed to be powered by various types of ship's power systems. It was expected that the receiver would be powered by +115vdc. Using accessories available from RMCA, operation on +230vdc was also possible. It was also possible to operate the receiver direct on 115vac but the receiver's power input section only provides a half-wave rectifier if AC operation is used. RMCA recommended that the MM-555140-B Isolation Transformer/DC supply be used to convert the ship's 115vac (or 230vac) to +115vdc for operation of the receiver. In fact, if the receiver was ordered with the matching table cabinet, then the MM-555140-B was included and installed inside the cabinet. If the rack mounted version of the receiver was ordered then the MM-555140-B was separate and was installed externally (somewhere in the rack.) Since the AR-8516 is AC-DC in concept, all tube filaments are wired in series. Most of the tubes are 3-volt filament types, e.g., 3BZ6, 3BE6, 3CD6, 3AL5, etc. but (2) 5U8, (3) 7AU7 and (1) 12CU5 tubes are also used. The SECTOR Control - The unfamiliar control on the AR-8516 is the SECTOR control which is part of the RF and Mixer 1 tuning in combination with the Crystal Oscillator. When the receiver is tuning from 80kc up to 4mc (first five bands) the RF and Mixer 1 circuits are bypassed and the Variable IF (acting as a RF amplifier, VFO and Mixer 2) tunes the receiver as a single conversion circuit. When tuning above 4mc, double conversion is used and the SECTOR control provides the correct tuning for the RF and Mixer 1 (and Crystal Oscillator) for each of the 2mc wide bands (4mc to 30mc.) Four sets of coils are used for ranges 4-12mc, 12-20mc, 20-28mc and 28-30mc. When combined with the selected Crystal Oscillator frequency the result is 2mc wide tuning ranges from 4 to 30mc. The "sector" refers to each 45 degree quadrant (A, B, C or D) of the RF/Mixer 1 variable tuning condenser's 180 degree rotation range. Selecting A, B, C or D will actuate a physical rotation of the RF/Mixer 1 tuning condenser rotor to the correct 45 degree quadrant required for tuning (plus the Crystal Oscillator) of the particular 2mc band range selected. The SECTOR switching is roller-chain coupled to the band switch. The RF/Mixer 1 tuning is gear-coupled to the main tuning control and the SECTOR selector knob will rotate as the TUNING knob is rotated. Although the SECTOR knob and RF/Mixer 1 condenser rotate with the main tuning, the correct quadrant selection is necessary and will require that the SECTOR knob be set to the "detent" in the correct position. The "PLACE SECTOR AT" indicator is illuminated and located in the viewing hole above the band switch. The main tuning dial, logging scale and meter illumination are also controlled by the SECTOR control in Bands 6 to 18. These lamps will only illuminate when the correct SECTOR position is selected. Serial Number and Cabinet - The serial number of the AR-8516 shown is SN:5830. The serial number incorporates the year of manufacture in the first two digits, e.g., "58" is 1958 and the receiver number is "30." There is also another tag mounted to the rear chassis that gives the month and year of manufacture, in the case of SN:5830, the tag reads "JAN 1958." The odd size panel height of 9.5" (rather than the standard 10.5") does limit the receiver's installation into generic-type cabinets without an obvious gap. Mounting the receiver in a rack will solve that issue. Of course the proper RCA cabinet does fit the receiver panel correctly. As stated in the manual, the AR-8516 was available as a table model receiver with the cabinet (with the MM-555140-B installed) or it could be ordered as a rack mount without cabinet (with the MM-555140-B as a separate accessory.) |
|
|
Antenna Connections per the Manual - There are four antenna terminals provided identified as A1, A2, GND and SHLD. The manual warns not to "ground" the SHLD terminal. This is because in Bands 1-5, SHLD is connected as part of the antenna input by essentially "shorting" the coax shield to the center conductor of a coax feed line. This allows the receiver to have a fairly efficient "T" antenna configuration in the LF and MW ranges. At Band 6, the band switch then connects SHLD to the receiver chassis allowing the use of a resonant, coaxial fed antenna. When using the typical unbalanced ham antenna, the center conductor of the coax is connected to A1, then A2 is jumped to GND and then just the shield of the coax is connected to SHLD. For the hook-up to work correctly, the antenna shield can't be grounded anywhere and must only be connected to the SHLD terminal. For hams the problem with this hook-up is for 80M (Band 5) and 160M (Band 4) operation. |
|
| A More Practical Antenna Hook-up - If you want to use a resonant unbalanced coaxial fed antenna for much better reception on 80M or on 160M, then connect the coax shield to GND, A2 also connects to GND and the coax center conductor to A1. Don't connect anything to SHLD. If an antenna tuner and tuned dipole antenna combo are used with this type of connection, then the tuner adjustments will "peak" normally. The ANTENNA (trimmer) won't function on Bands 1 thru 5 because the antenna signal is going to the Variable IF (the receiver is single conversion in Bands 1-5.) |
|
Building a Replica MM-555140-B |
| The AR-8516 manual states that if a very low hum on the audio output is desired or necessary then the receiver should be operated on +115vdc. I decided to build a MM-555140-B since finding an original would probably take a considerable amount of time. Luckily, the AR-8516 manual has the schematic for the MM-555140-B and there really aren't any unusual parts required for its construction. In fact, everything to build my homebrew version came from the various junk boxes I have around here. I used a 150VA 1:1 Isolation Transformer as the main component. The receiver requires about 80 watts for operation, so there's lots of reserve capability in the transformer. Other components were a full-wave bridge rectifier, two 40uf 300vdc electrolytic capacitors, a 22K 2W CC resistor, a 25 ohm 50W WW resistor and a 6.5 ohm 60W WW resistor. The two large dissipation wire wound resistors were made up of a parallel combo for the 6.5 ohm 60W (15+15+75 ohms) and a parallel combo for the 25 ohm 50W (two 50 ohm 25W resistors.) >>> |
 |
|
|
AR-8516 SN:5830 Refurbishment |
| When acquired, SN:5830 did turn on and did receive some
signals, but it wasn't fully functional and really wasn't
performing at its design capabilities. Initial inspection found
five of the 18 tubes
were bad with two showing significant leakage (shorts.) The AF
gain pot was internally damaged and no amount of cleaning would
fix it since the carbon track was cracked in several places.
Luckily, I had a NOS 500K A-taper (audio taper) A-B pot to replace the bad original. De-Oxit
was used on the Bandwidth switch which helped it to function
correctly. The receiver was the rack-mount version so it didn't have
a cabinet and for some reason RMCA never offered a dust cover.
I guess they figured that the rack console would protect the
receiver chassis. However, decades of storage in a greasy
environment certainly had the chassis looking terrible. There
was a little minor corrosion but most of the contamination was
dirt, waxy-grease and probably tar-nicotine. The front panel had its share of
"battle-scars," "rack rash" and even some green paint
splatter. The knobs were just plain "grungy" with crud that
seemed waxy-like in all of the flutes and recesses. There was white residue
in the front panel screw slots implying
that the panel and knobs had been
rubbed-down with polish or wax in the distant past.
Thorough cleaning required major disassembly. The knobs were soaked in Dawn dish soap for two hours to loosen the waxy grunge after which they were scrubbed with a short-bristle brush to thoroughly clean. The knobs were then rubbed-down with a dry cotton flannel cloth to slightly polish the surface finish. The RCA meatball badge needed to be dismounted in order to remove the pitting (600 grit AlOx paper,) cleaned with Glass Plus along with a wooden toothpick used to "dig-out" the wax in the recesses between the letters and finally the badge was polished with Wenol's (British equivalent of Semichrome.) The chrome handles had minor pitting that was removed with Glass Plus and 600 grit AlOx paper followed by polishing with Wenol's. The grunge on the chassis was removed with WD-40 and an acid brush (with the bristles cut short.) This was followed by Glass Plus to remove the WD-40 residue. In some stubborn places, isopropyl alcohol would cut the grime better. The plastic dial cover was removed, cleaned and polished. The meter scale, the sector scale and the kilocycle/logging scale were made of photosensitive phenolic that had darkened but careful cleaning with Glass Plus brightened up the yellow-amber color significantly without damaging the nomenclature. The dial scale was a metalized over-lay applied to the drum. It cleaned up easily because the material and the nomenclature were particularly durable and could be cleaned a bit more aggressively than expected. I installed "matching" tube shields. These were the standard flat aluminum type because the IERC-type of heat-reducing tube shields are really only necessary when a receiver is going to be operated 24/7. I matched the original paint on the front panel (using 5 colors of Testor's model enamel - gray, black, white, dark blue and brown) and touched up most of the "battle-scars." The silver dial bezel was touched-up using thinned Testor's silver paint. A flat-razor blade was used to remove the green paint drops. The panel mounting screws were cleaned and painted black before reinstallation. After cleaning the tops of the two shield covers (RF-Sector condenser and Variable IF condenser) they both look silver in the photos. Actually, they appear slightly gold tint, though much lighter than the chassis. The gold finish is called iridite and it was a chemical bath that colored the sheet metal during manufacture and before assembly. |
| Disassembly issues NOT
mentioned in the manual: The BFO pointer is
dual set-screw mounted to the outer coaxial shaft of the vernier BFO
adjustment. To remove the front panel this pointer has
to be dismounted but you can't get at both set screws from the top. The
right side panel has to be removed to access the set screws to
dismount the pointer to allow front panel removal. All of the set
screws in the knobs and couplers are Bristol (spline) types. There
are four shaft bushings, two of which have "C" clips associated
with them (ANTENNA and TUNING.) The Antenna (trim) coupler set screws are easy to
access from the top. But, the TUNING shaft will require removal
of the "C" clip to be able to dismount the front panel. Also,
three of the bushings are standard size but one is much longer
than the others. The long bushing is used for the TUNING shaft for better support of the shaft which does have a large
counter-weight installed on it. The dial string assembly rail for the dial pointer
is mounted to the front panel. When the four screws are removed
this piece can then "flop around" and rub against and mar the
dial scale. Place several paper towels over the dial scale to
protect it during disassembly. Note that the PHONES jack is
isolated from the chassis and panel by two shouldered fiber insulating
washers. This jack should not be contacting the panel or
chassis (problems could result if using direct AC line operation.) Removal of the dial bezel requires removing the front
panel first, then plastic
dial cover piece is dismounted from the rear, then the bezel is dismounted from the
two retaining strips and then the two strips are dismounted from
the back of the panel. This assembly can't be installed or
removed if it is all put together, it has to be installed (or
removed) by dismounting each individual piece. Sea Trials - Dec 12-15, 2020 - After reassembly and completion of the DC power supply, I operated the AR-8516 for a few days to get the "feel" of the receiver. As acquired, this AR-8516 had already been meticulously and carefully re-capped so I was pretty confident that the receiver could operate for long time periods without any serious problems. The receiver seemed to function pretty well but it was obvious that some improvement should be possible. The next day I found a very minor problem that had a significant effect on reception. The jumper between A2 and GND had been a piece of solder (the lazy tech's jumper) and it had broken at the A2 terminal. I replaced the solder with a TC jumper. The signal reception level was greatly improved. I tuned around 12.965mc and copied one of the Seoul, So. Korea coastal beacon stations, HGW2. Also, a little lower in frequency and one of the Chinese coastal beacon stations, XSQ was heard. The next morning, I tried 40M and some of the SSB signals were intensely strong. I tried HGW2 on 12.965mc (should be at 12.915mc,...more on that problem in the alignment section) and it was very strong. Fire Drake, the Chi-Comm jamming signal on 9.85mc (it changes frequency all the time since it's jamming Radio Taiwan which also changes frequency all the time to avoid Fire Drake,) was very strong and showed about 25db over 1uV on the carrier level meter. Trenton Military Aviation Weather (Trenton, Ontario, Canada) on 15.031mc (should be 15.035mc) was heard with a very strong signal. One other thing noticed before, with the broken jumper, I couldn't seem to find a good match with the antenna tuner. With a jumper installed, now the tuner showed a very positive increase in reception when the antenna match was "tuned." Also, the ANTENNA (trimmer) on the receiver would "peak" now. It was also noted that the best results on SSB or CW was to reduce the RF Gain and advance the AF Gain which is typical of a receiver without a product detector. If the RF gain is advanced too far it seems to "mask" weaker CW signals with excessive background noise, again typical of a standard diode detector. In AM Voice, the RF gain can be fully advanced and the AVC allowed to control sensitivity. The RF tracking is fairly close with the error at 15.000mc being -6.0kc. The 500kc oscillator is off by about 400hz and this affects 45kc IF and the 1.5kc-800-100hz bandwidth performance (which barely worked) and the 500kc calibration marker accuracy. Although performance seems pretty good, a full IF/RF alignment will assure that the receiver is operating at its design specs. |
|
| Return to the
Workbench - Dec 18, 2020 - It's pretty much standard procedure
to perform an initial refurbishment to get a set going, then use
it for a while to assess how it's going to work out and then plan what
else needs to be done to complete the rebuild. Although
performance was generally good, and it seemed likely that the
AR-8516 could do a very good job as a station receiver, there
were other mechanical and electronic things that had to be done
to enhance the receiver's functionality. Here's the list:
1. Mechanically aligned the slide rule dial to more accurately agree
with the kilocycle dial. Pre-alignment necessity. To say the alignment procedure is lacking details is an understatement. I found it necessary to make lots and lots of pencil notations about important component values, test point locations and even drawings of the IF transformer connections. The 455kc IF transformers pin-outs are important because the primaries or the secondaries or both have to be "loaded down" with 10K resistors during alignment. Luckily, the IF transformer pin-outs are vaguely mentioned in one of the schematic notes,...no where else,...and even that information is incomplete requiring a visual and continuity check to verify. NOTE: IF transformer pins viewed from the bottom, green dot is pin 1, go clockwise for pins 2, 3 and 4. Next, the RF signal generator input for IF alignment is to the grid of V105 Mixer 1,...no problem,...except that the bottom of the V105 socket is located within a completely shielded compartment (with more than thirty 4-40 screws mounting the cover.) I used a seven-pin tube test extender socket to input the signal to the V105 grid. >>> |
|
| >>> Another complication in the IF and in the RF tracking alignment process is that the receiver has
to be on its side since there are adjustments both on top of and
underneath the chassis along with having to observe the front of
receiver for the dial readout. I had to use a 4"x6" hand mirror
to verify tuning dial readouts and sometimes the locations of
various trimmers to keep from bending over or leaning around the
side of the receiver for every adjustment. I guess I could have
turned the receiver around so the panel was in front but
then all of the difficult to see small chassis-located
adjustments would have to either require a mirror and
bending over and around the chassis or some other
uncomfortable contortions. I think that at least a couple of days of reading, making notes, visually locating the important component junctions and getting all of the resistor loads, coupling caps and alignment tools together will help to ease the alignment process. |
| IF Alignment
- This starts with the 455kc fixed IF. The only unusual
requirement is the loading down of some of the IF transformers
primaries or secondaries
with 10K resistors. This is pretty easy to accomplish if the 10K
resistors are on short flexible wires with small alligator clips
on the ends. The bottoms of the IF transformers are easy to
access for attaching the 10K loads. Two 10K loads are required
because T124 and T125 have both primary and secondary loaded
simultaneously when doing the adjustment. There is a trimmer
capacitor for the
mechanical filter that is adjusted for a relatively equal response between
the 6kc bandwidth and the 3kc MF bandwidth. The 455kc BFO is also
adjusted at this point in the procedure. The 45kc IF does require a signal generator that will produce RF at that low of a frequency. I normally use the HP 606B for alignments but its lower limit is 50kc. Luckily, I also have an excellent General Radio Type 1001A RF generator that will produce a RF signal as low as 5kc. For the 45kc IF, I had to use the GR 1001A for the signal source. Four torriod type LCs are adjusted with trimmer caps and one transformer L is adjusted. The 45kc BFO is adjusted at this time. The Variable IF is adjusted to track perfectly (or as close as possible) from 2.0mc to 3.9mc. This is accomplished with adjustable L and C and referenced to the kilocycle dial for maximum accuracy. This completes the IF alignment. |
RF Tracking -
The single conversion low frequency tuning is aligned first.
This requires adjusting six L and C adjustments per band for a
total of 30 adjustments. Only the 2-4mc single conversion band
(Band 5) can be aligned using the kilocycle dial. Bands 1-4 are
adjusted using the slide rule dial. After the tracking is adjusted on
Bands 1-5, then the 45kc wavetrap is adjusted at 80kc before proceeding
on to the double conversion RF tracking alignment. Bands 5-18 are double conversion and are aligned in four major divisions, 4-12mc, 12-20mc, 20-28mc and 28-30mc. Each Sector A, B, C and D requires Crystal Oscillator peaking and RF-Mixer adjustments. Seven crystals are utilized but six are doubled and used on harmonics to allow the 13 bands to operate. There are 23 adjustments plus peaking the ANTENNA. I couldn't get "to the kilocycle" accuracy using the kilocycle dial. It's pretty close and does meet the specification of dial accuracy of "within 10kc" which seems to be pretty good (12-14mc is the exception - covered in the next section below.) For frequencies below 4mc, dial accuracy spec is 0.5% which is very easy to maintain. The 500kc oscillator is adjusted using WWV and the beat note. Since the BFO is on when in the 500kc CAL position, the BFO should be set to zero so that only the 500kc oscillator and WWV are heard when listening for the heterodyne zero beat. There's also two adjustments for the AF feedback that are adjusted for minimal "ringing" when in the narrow 100hz bandwidth. Over 75 different adjustments are necessary for a complete AR-8516 IF/RF alignment. |
| Manual Errors and Suggestions - There are two obvious errors or conflicting data in the manual's alignment procedure. T121 is identified as a 12mc adjustment in the manual's adjustment drawing. The receiver silk screening and the manual written procedure are correct indicating that T121 is for a 2.0mc adjustment. The second error is T115 is correctly identified in the adjustment drawing as 220kc and the receiver silk screening also shows the correct 220kc. The alignment procedure "frequency table" incorrectly calls out 210kc for the adjustment frequency. Some of the instructions are confusing in the wording used and require further checking to confirm what actually is required. The use of the IF 10K loads is a good example of confusing language. As a matter of poor layout, the manual has two chassis illustrations dividing the written alignment instructions. My manual was in a binder and the pages could be taken out as needed. If only a bound manual is available then print copies of alignment instructions so the book pages aren't worn out from flipping back and forth as the alignment proceeds. |
| Post-Alignment
Performance and Issues - No doubt, the alignment
significantly improved sensitivity and certainly got the 45kc IF
working correctly along with the Bandwidth 1.5kc, 800hz and
100hz working very well. Listening on 20M quickly turned up a
ZS6 amateur from South
Africa. Trenton Military out of Ontario, Canada was easily
found at 15.035mc since their signal was very strong now and the
dial accuracy noticeably improved. WWV 15mc was about 1.5kc off frequency readout and WWV
10mc
was about 1kc off. However, WWV 5mc was off about 10kc (still
within the factory specs.) 40M hams required reducing the RF Gain down
to 4 for the very strong signals. 80M performance was quite good and the tracking on that
band (Band 5) was easily within the 0.5% spec. Just a
note,...there's no noticeable difference in using either a 3.2Z
or a 4.0Z speaker (as expected.) So, that was the good news,...but two problems were noted.
12-14mc Band Tracking - For this problem I don't think there's any easy solution. The 12-14mc band tracking is off by about 45kc. Since all other bands are very close in their tracking, I would think that the Crystal Oscillator is the problem with the crystal being the most likely suspect. The crystal is Y103, an 8.00mc crystal and it is used twice, on 12-14mc and also on 28-30mc. A test on 29.0mc revealed an error of about 90kc which seems to confirm that the crystal Y103 is the likely problem. From what I've heard, there are virtually no companies that supply newly made crystals anymore. I checked to see if the R-390A happened to use that frequency crystal but it doesn't. In addition to finding an 8.00mc crystal there's also the physical access to the crystals in the receiver which is beyond difficult. Looking at the photo to the right, the crystals are soldered directly to the band switch (they have a green coating on the metal housing.) I looks like replacing a crystal requires extracting the entire Crystal Oscillator as a unit. It's not a module but note in the photo that the band switch shaft uses a slotted nylon Oldham-type coupler which implies the Crystal Oscillator is removable. Since performance on the 12-14mc band is excellent in all other areas, I'll probably just have to accept that frequency readout on that band (and on 28-30mc) is off. Even with a 45kc dial readout error,...at 12mc the accuracy is still better than 0.5% (0.5% is 60kc at 12mc.) Note: Used 8.000mc crystals are available from Alltechs in California - seen on eBay - about $5 plus shipping. 200kc to 500kc Band Sensitivity - The second issue was poor performance on the 200kc to 500kc band (Band 2) and this was easily repaired. Low sensitivity on this band was due to the misalignment of VFO at 220kc. I must have initially set the RF gen to 210kc as indicated in the manual alignment instructions but actually had the receiver tuned at 220kc as indicated on the receiver chassis silk screening (after all, I was looking at the receiver dial in a mirror.) To correct the problem, obviously I realigned with the receiver tuned to 220kc and the RF generator also tuned to 220kc. The other alignment adjustments (six adjustments, three for 220kc and three for 500kc) were also peaked for Band 2 and afterwards the sensitivity was quite good and the tracking was excellent. In the evening, I set up the AR-8516 with the Pixel Loop and tuned in several NDBs in the 300kc to 420kc part of the spectrum. Best DX was LLD 352kc in Hawaii, several Canadian NDBs in Manitoba and Saskatchewan along with several Mid-west 25W NDBs. With a wire antenna, signals were stronger but so was the noise which is normal. Dial accuracy was good but the dial's scaling is fairly vague. Dec 23, 2020 |
|
| Performance as a Station Receiver - I set up the AR-8516 with the Collins 32V-3 as a station to operate on 75M for the MRCG net and the Vintage Military Radio Net. A Collins 270G-1 speaker was used as the audio reproducer. The receiver doesn't have a remote standby circuit so the front panel standby position of the operation switch has to be used. Also, with the 32V-3 set up I use a vacuum relay for the T-R operation for isolating the receiver antenna input. Since 75M is tuned in on Band 5, the AR-8516 is operating as a single conversion receiver and because the signal routing is to the Variable IF, the ANTENNA (trimmer) isn't functional. However, using a tuned antenna (~50Z for the transmitter) will be a close match to the receiver input Z. Frequency readout is primarily on the slide rule dial where the accuracy seems reasonable. Sensitivity on 75M really isn't ever too much of a problem for just about any receiver. Selectivity is much more important. 6kc bandwidth can be used most of the time but it's nice to also have a 3kc mechanical filter for tough QRM. I run the AR-8516 on +115vdc and there's no hum at all on the audio output. It's very clean sounding and voice articulation is reproduced quite well. The result is easy-to-understand AM QSOs (most of the time.) The only complaint is the lack of a remote standby actuated with the PTT. >>> |
| >>> For the most part, any shortwave transmissions
are easy to receive. CW also is reproduced nicely. SSB uses the
adjustable BFO position to select either USB or LSB and
selecting the "SSB" position on the operation switch
increases the BFO output to improve the signal to BFO injection ratio. Stability in the CW or SSB mode is excellent. Using
the narrow Bandwidth will depend on how well the receiver is
aligned. When first tested, this AR-8516 hardly functioned in
the narrow Bandwidth positions since alignment of the 45kc is
crucial for these functions. After alignment, the narrow
Bandwidth selections function great and are a real benefit for CW in
crowded band conditions. Since there's no crystal filter, the
operator has to rely on narrow bandwidths to eliminate QRM. Also, the narrow bandwidth selections
utilize the 3.1kc mechanical filter at 455kc and then use the
45kc IF to provide 1.5kc, 800hz and 100hz bandwidths. When
properly aligned, the AR-8516 provides excellent selectivity
options. Update: May 2021 - I built a wooden cabinet for the AR-8516. Nothing fancy. In fact, I built it out of "used plywood" and from some of the edges, I guess that's pretty apparent. But, it is square and it does a good job for having a place for the loudspeaker and other accessories to set. I painted the cabinet gray and I designed it to conform to the dimensions of the front panel and the full depth of the chassis. Overall, the gray paint finish helps to disguise the fact that the cabinet is made out of wood (well,...maybe a little, especially in average room light,...with stuff piled on top.) I think it's a nice improvement to the AR-8516 "looks" overall and certainly provides more bench space than the open chassis did. I've been using the AR-8516 paired with an ART-13A transmitter and have found it to be an excellent station receiver with very nice audio and excellent selectivity. The lack of "remote standby" can be compensated for by using the front panel standby switch. With the generous band and frequency over-lapping, the military radio nets - all being at the upper end of 75M - can actually be tuned in on Band 6 (4-6mc) to take advantage of the double conversion. Although not shown in the photo, I'm still using the Collins 270G-1 loudspeaker with this receiver. |
|
|
Facelift for the AR-8516 Homebrew Cabinet |
|
|
UPDATE: Feb 22, 2025 - First, a few
Mechanical Problems -
About six months ago I went to use the AR-8516 and it seemed to work fine.
It had been in operation for about 30 minutes when suddenly the audio
was reduced by about half. I didn't check anything at that time but just
put the receiver back into the upstairs storage area. Since that was
about six months ago, I
thought I should probably take a look at the AR-8516 and see what had
happened. This time when powering it up, the receiver came to life with
no problems. But, as soon as I moved the ANTENNA control, there was a
lot of "scratching" noises and a reduction of signal strength. I found a
spot on the ANTENNA adjustment where everything returned to normal. The
antenna trimmer C is located inside a shielded housing with the RF/Mixer
variable, so that had to be removed. I checked the antenna trim C with a
DMM and there was a positive indication of a short between the plates. A
close examination with a bright light revealed that the rearmost rotor plate
was slightly bent and was shorting to the adjacent stator plate at one portion of the rotation. I
very slightly bent the rotor plate until I had no indication of a short
during complete rotation. I cleaned all of the contacting surfaces
involved with a small amount of DeOxit and then replaced the shield
cover. A test of the receiver and that problem was eliminated and now
the ANTENNA adjusted correctly with no noise. I have no idea how this
trimmer rotor plate got bent. I know I didn't have this problem in 2021
or I would have fixed it since it was a simple repair. Weird, since the
trimmer is in a shielded compartment. Next was the BFO and this was another simple mechanical problem. The BFO knob has an internal collett-type of grip that won't mar the shaft. The BFO shaft is actually a vernier mechanism between the knob-shaft and then, behind the panel, the vernier output via a coupler to the BFO assembly. I removed the knob and then could see that by rotating the shaft directly with my fingers, the BFO could be adjusted without any problems. The knob's set screw was a spline-head Bristol screw that had been severely worn in the spline socket. Apparently I hadn't tightened the screw enough so the collett didn't grip the shaft and that was causing the BFO adjustment to be erratic as the knob "slipped" on the shaft. >>> |
| >>> Using a very good condition spline wrench, I was able to get the wrench
deep into the socket and that allowed tightening the screw sufficiently
to prevent slipping. Tested the BFO and it was set correctly for 0
equaling "zero beat." Again, unknown why this problem seemed to surface
four years later. I'm sure the BFO operated fine back when I refurbished
this receiver. I suppose
it's possible that expansion and contraction might have loosened the collett screw over a four year period (although the receiver has been
inside the house for that time period.) Facelift Plan - I'm going to complete the wooden cabinet for the AR-8516. To cover all of the raw edges and visible screws and seams, I'm going to install flexible "peel and stick" corner molding. Just the top and bottom edges and then the front edges will have this trim installed. The size and shape of the trim will cover all screws and all of the defects visible. I'm going to add large diameter stick-on felt feet to make the receiver cabinet easier to slide on the table top (the AR-8516 is a very heavy receiver.) I'm going to add panel-rack screws to secure the receiver in the cabinet. March 5, 2025 - Cabinet trim is finished. I found some 3/4" x 3/4" "peel and stick" Chrome trim. It's a PVC base with adhesive on the inside and chrome on the exterior. It's super-easy to work with since it only has to be marked for dimensions and then it can be cut with scissors. I had to measure and pre-cut each joint for a 45º angle to fit the corners properly. Per the instructions, I used a heat-gun to preheat the wooden surface and the strip to be applied. After installation, the protective plastic covering could be removed. The AR-8516 was then installed into the cabinet and then the screw holes drilled for the rack mounting. The screws self-thread since they are going into wood. Large felt pads were installed on the bottom of the cabinet. The chrome trim has improved the appearance of the cabinet quite a bit. Of course, it doesn't look anything like the original metal table cabinets that were used for this receiver or for the CRM-R6A. But, those cabinets are extremely rare and are never found separated from their receivers. |
|
| References: Hard Copy: 1. Manuals - Originals or copies,...it doesn't matter as long as the information is there. Original manuals for AR-8711, R-203A/SR (TM11-875) and the AR-8516. Copies for IP-501-A, AR-8503, RAZ-1 and the AR-8510. Manuals are necessary for detailed information on the circuit including the schematic and alignment instructions. Manuals don't always provide complete information but they are better than no information at all. 2. IP-501-A - There are a few write-ups from as far back as Popular Electronics in the 1960s up to Radio Age and Antique Radio Classified articles. AWA OTB has some information. These write-ups have general information. 3. Other RMCA gear - Very little detailed information on specific RMCA receivers. Mostly general information. 4. RCA and RMCA History - Most RCA formation history is from "Keeping the Stars and Strips in the Ether" by Cmdr. Stanford Hooper, who was one of the Navy participants in the American Marconi purchase. One note is that Hooper refers to United Fruit Company as the Tropical Radio [Telegraph] Company, the company that United Fruit Company used for the operation of their wireless stations. This article appeared in Radio Broadcast magazine June 1922 issue and contains all of the details of the creation of RCA from the American Marconi purchase by GE. Also, Alan Douglas's "Radio Manufacturers of the 1920s" RCA section in volume three has details on the creation of RCA. Online Material: 1. Inland Marine Radio History Archive - has the most RMCA information I could find. Lots of photos of RMCA equipment and history links to other websites. URL is www.imradioha.org/rmca.htm 2. Maritime Radio Historical Society - San Francisco area based, has a lot of info on RMCA coastal station KPH and the equipment that was used there. URL is www.radiomarine.org |
|
|
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 © 1997/2025
Under the chassis showing the bee's
wax impregnated coils.
