Radio Boulevard
Western Historic Radio Museum
 

Navy Department - Bureau of Ships

RAO Series of Radio Receiving Equipment

National Company, Inc.  -   Wells Gardner & Company

History of the "Moving Coil" aka "Coil Catacomb" Design
 Detailed Information on all RAO Versions
Navy RCX Panoramic Adaptor with RAO-7
Serial Number Log
Detailed Restoration Write-ups for RAO-3 and RAO-7
Expected Performance Information

by: Henry Rogers WA7YBS

RAO-7 - B&W artwork from NAVSHIPS 900,356

History of the National Company, Inc. "Moving Coil" or "Coil Catacomb" Design

Many radio engineers of the thirties firmly believed that the absolute best sensitivity and stability of a communication receiver's RF amplifier, First Detector and Local Oscillator could only be achieved by using "plug-in" coils. This type of approach eliminated problems of lead length, shielding and stability along with isolation of unused tuned circuits - problems that were commonly found in broadcast receivers using conventional rotary bandswitches. Plug-in coils were a hassle, no doubt. Handling three individual coils for each band change, storage of the unused coils and how to remove B+ when changing coils all added to the counter-belief that plug-in coils were archaic. James Millen, National Co.'s General Manager and Chief Engineer, was one of the designers that insisted the best receiver performance was achieved using plug-in coils. But, how to effectively eliminate the valid arguments against plug-in coil use in a new product?   

Certainly, National was having fabulous success with the HRO receiver, which had been in production since early 1935. It was well-known that the HRO's legendary performance was in-part due to its plug-in coil sets. The HRO wasn't hassle-free though and Millen realized that for the SWL (Short Wave Listener) and intermediate-level hams, in other words, those who didn't have the experience or couldn't afford the $200+ HRO receiver, there had to be a design that would provide the excellent performance of plug-in coils without all of the hassles and expense.

The mechanical action simulated plugging in a three coil set for each band with the ease of turning a knob while keeping all of the unused coils isolated and shielded. "Switching noise" was eliminated by routing the RF and IF amplifier screen voltage through the foremost pin split-contacts of the LO coil section. The split-contacts were not soldered together but rather had the screen voltage wires connected to each of the two halves of the contact and when the coil pin, which wasn't connected to anything inside the coil catacomb, engaged in the two halves the circuit is completed and the screen voltage was then routed to the RF and IF amplifiers (changed later to RF/IF cathode connection.)

The initial NC-100 receivers used a PW gear drive and a PW-D micrometer dial (PW-D was National's identification of their "micrometer" dial.) Since the NC-100 was intended for the SWL or intermediate-level ham, a different type of signal strength or tuning indicator was utilized. Since a cathode ray tuning indicator was part of the NC-100 design it was going to require strong and consistent AVC voltage. This required the incorporation of an Amplified AVC stage in the receiver. Since nearly all hams were running on CW in the thirties and their receivers were always operated with the BFO on and the AVC off, the use of an amplified AVC stage implies that National was designing the NC-100 more for voice reception of AM signals from Shortwave Broadcasters. IF is National's standard 456kc. 12 tubes were used in the NC-100 circuit. The NC-100 tuned from .54 to 30 mc in five tuning ranges.

Finally, there was the Push-Pull Audio output stage using a pair of 6F6 tubes driving a 10" Rola Type K-10 electro-dynamic speaker. Even an optional 12" electro-dynamic speaker was available. Again this great audio section implies the reception of voice and music, not the "dots and dashes" of International Morse. Since the intended market for the NC-100 was not necessarily hams, the receiver was introduced with a fabulous "art deco" front panel that featured geometric-linear black panels and black octagon control nomenclature layout combined with the natural aluminum finish along with a red highlighted central panel for band in use indication. The PW-D combined a bluish-gray index dial with a red number dial along with a green pilot lamp and the green "eye tube" (when in operation) finished off the striking "art deco" design of the receiver. The NC-100 and the NC-100X were introduced in August 1936 and were available from dealers in September 1936, with the initial pricing at $105 and $127 respectively.

These early version NC-100 and NC-100X receivers were soon followed by the "ham bands only" version, the NC-101X. The ham band versions tuned 160M, 80M, 40M, 20M and 10M. Still, the micrometer dial was retained requiring the operator to use some sort of correlation between the tuned frequency and micrometer reading. That correlation took the form of tuning graphs that were included in the operating manual.

photo above: NC-100XA SN: 58-R, ca. late-1938. The articulated dial pointer indicates that Band B is selected and that the receiver is tuned to approximately 11.85mc.
 

The NC-100A

The use of an indirect-readout device such as the PW-D probably accounted for more than a few complaints to National from casual users who had to constantly refer to the manual for a graph that provided frequency versus dial readout correlations. The competition's direct-readout dials also could have been a factor that resulted in National revamping the entire NC-100 line to replace the PW-D micrometer dial with a direct readout, illuminated tuning dial. The suffix "A" was added to all NC-100 receivers that had the new dial installed. The introductory ad appeared in the June 1938 issue of QST. For sometime, the NC-101X was available with either the micrometer dial or with the "A" version dial as the NC-101XA. After all, the NC-101X was a ham receiver and many hams enjoyed the correlation exercises involved with using the PW-D.

One of the interesting features of the new "A" version dial was its articulated pointer. When the band was changed, the dial pointer would automatically increase or reduce its apparent length so that its red tip would line up with the selected band's tuning scale, thus indicating the "band in use." This required the pointer "lifter" mechanism to track the tuning while maintaining the proper length of the pointer. A dial cord that was anchored to the band changing shaft and then routed via a pulley system to pivot against the tuning condenser drive shaft so that the action of tuning the receiver didn't affect the apparent length of the pointer while it tracked around the dial. The pointer-lifter mechanism and the entire articulation system seemed overly complex for the simple task of a band select indicator. Small wonder that during WWII this feature was eliminated and replaced with a gear-driven indicator dial mounted to the band change shaft. A logging scale was included to allow for precision frequency resetability that was comparable to the PW-D dial's accuracy. Initially, the dial cover was a pane of glass. Most later military versions had the glass replaced with plexiglass.

In addition to the new tuning dial, the cabinet itself was increased in height from the nine inch height of the NC-100 to ten and a half inches of the NC-100A. The eye-tube was replaced with a "behind the panel" S-meter that was installed on all versions now. A push-pull switch allowed disabling the S-meter when CW was being received, just like the HRO receiver. On the earliest versions, the white scale S-meter was installed but these were soon replaced with the yellow scale meters. With the S-meter installed, the total tube count of the NC-100A circuit was reduced to eleven tubes since the cathode ray tuning eye tube was no longer used. >>>

The Tone control on early NC-100A versions used a large inductor in the circuit. Push-Pull 6F6 tubes were used in the audio output with the output transformer located on the external speaker. The speaker included was a 10" diameter Rola and the speaker-cabinet assembly was designated as MCS-10.  

With the NC-100XA, some of the control locations were moved with the addition of the Crystal Filter (still at 456kc.) The S-meter switch was moved from adjacent to the meter to the lower part of the panel. Different nomenclature was required so the small panels were changed to reflect the additional controls needed on the "XA" version. The selling price for the NC-100XA was $147.00 from Allied Radio in 1939. Earliest reported "A" version is a NC-101XA with SN 130-M which would date from around May or June 1938.

Throughout the NC-100A production, the circuit went through several minor changes. Later in production, an adjustable Noise Limiter circuit was added (Nov. 1939.) The NL circuit changed the 6C5 second detector to a 6C8 duplex triode to provide the second detector function and the Noise Limiter function (keeping the tube total at eleven tubes.) The 6J7 AVC Amp was replaced with a 6F8 duplex triode to provide AVC amplification and also to add an audio preamp (1st AF Amp.) Also, with the NL addition, the location of some of the controls was rearranged. The Audio Gain control was changed to a standard grid input on the new 1st AF amp and the Tone control was changed to an RC type control which eliminated the audio choke that was used in the earlier versions. Some of the NC-100A and NC-100XA versions featured a "weighted" tuning knob that added a fly-wheel affect to tuning the receiver. The weighted knob wasn't used on the military versions. Production of the NC-100A and NC-100XA probably lasted until the NC-200 was fully in production (Oct. 1940) but, by that time, National was working with both the Navy and the USCG to supply them with NC-100A receivers that were modified specifically for their needs.

photo above: NC-100XA  SN: 58-R  Top of chassis. Note that only one filter choke is used since the field coil of the loudspeaker doubles as a second filter choke.

RAO and other Military Catacomb Receivers - Production Begins

U.S. Navy RAO -1 -  The contract for the RAO-1 is dated Dec 11, 1939. The contractor is General Electric Supply Corporation in Washington D.C. with National Company being the manufacturer. National began supplying the US Navy with the initial, single-preselection RAO and RAO-1 probably sometime in 1940. The double preselection RAO-2 was built on a contract issued April 30, 1942.

U.S. Coast Guard R-116 Receiver - The R-116 double pre-selection receiver was built on a contract dated 15 May 1941. The R-116 has a lot of the later RAO-2 characteristics but it also has several major differences. For example, the R-116 uses a separate power supply with both the power supply and the receiver in a rack mount configuration. Additionally, the receiver actually has six tuning ranges covering frequencies from 1.5mc to 27mc. With six bands and less total frequency coverage than the NC-100XA or RAO, the R-116 tuning rate has much more "spread" and makes tuning in CW stations easier. The coils used in the R-116 are wound on ceramic forms where the RAO used R-39 material for forms. The catacomb used was from the NC-200 with its 18 coil compartments. But the R-116, with the addition of the six 1st RF coils in the smaller catacomb, brings the total coils used in the R-116 to 24. Due to the second catacomb and the total length of the dual catacomb system the rack gear was moved from the front of the catacomb to the center of the larger catacomb and a second rack was installed at the rear of the smaller catacomb. The band switch knob operated a gear drive and a long shaft with pinion gears that engaged the catacomb racks and allowed band switching. In the RAO and RBH receivers, the system parts were somewhat simplified though the drive remained basically the same. The R-116 layout of controls is quite different from the RAO, the dial escutcheon is reduced in size and several different types of vacuum tubes are used in the circuit. The Crystal Filter only has two positions. The R-116 was normally rack mounted using a "picture frame" adapter that bolted to the corners of the receiver cabinet. The R-116 is possibly the first "catacomb" receiver with the double pre-selection modification since its contract date is May 15, 1941 and the RAO-2 (first RAO with double pre-selection) contract is dated April 30, 1942.

U.S. Navy RBH Receiver - The RBH was built for the Navy almost concurrently with the RAO. It's very similar to the RAO but had a different frequency coverage, that being 300kc to 500kc and 1.7mc to 30mc. The RBH utilized a 1500kc IF that then allowed continuous tuning from 300kc to 500kc. The RBH followed a similar evolution as the RAO in that the initial RBH was single preselection while the RBH-1 and later versions were double preselection to reduce LO leakage. The National RBH was also identified as the NC-156 with the Navy identification CNA-46114 built on a contract dated June 28, 1941. The later RBH-1 and subsequent versions were identified as NC-156-1 with the Navy identification of CNA-46188.

Note that the proceeding receivers had their initial contracts issued before the USA became involved in WWII.

U.S. Army Signal Corps NC-100ASD - The Army also wanted a NC-100-based receiver and in a 1943 contract was supplied with 1000 NC-100ASD receivers. These receivers are very close to the civilian NC-100A and even have the articulated dial pointer. The push-pull audio and field coil speaker circuits were changed to PM speaker and single-ended 500Z ohm audio. The Army found little use for the NC-100ASD and there was a large quantity available post-WWII. Newark Electric sold the NC-100ASD for $115 and that included the matching loudspeaker and a manual that was serialized to the receiver (probably removed from the original crate when sold.) There is some evidence that a Signal Corps AN/GRR-1 was built that was an Army version of the NC-100XA but no known examples have turned up. Some NC-100ASD receivers were used in Army DF equipment.

More R-116 Details - The R-116 lacks an internal power supply requiring that the receiver operates from a separate AC power supply (rack mount dual supply) or from a DC source such as batteries, motor-generator/battery combo or other types of shipboard power sources. Voltages required are 6 volts for the tube heaters and +230vdc for the B+. A four wire power cable exits the rear of the receiver. I used a National 697 "dog house" power supply as a power source. Audio outputs are 300Z and 600Z. Serial number on the R-116 shown is 46.

Tubes Used - RF1 - 6K7, RF2 - 6K7, Mix - 6K8, LO - 6J5, IF1 - 6K7, IF2 - 6K7, Detector - 6J5, AVC Amp/Rectifer - 6SF7, BFO - 6SJ7, 1stAF - 6J5, AF OUT - 6V6,

Components - Photo to the left shows the front panel of the R-116. The dial escutcheon is reduced in size and the controls use small round nomenclature plates. The photo right shows the chassis of the R-116. The circuit below the tuning condenser (in the photo) is very similar to the NC-100XA or RAO with the exception of the oil filled bypass capacitors that are mounted to the chassis. Above the tuning condenser, where the AC power supply normally is on the RAO or NC-100XA, are several unusual components. The large box is an audio filter that can be switched in from the front panel. The left choke is also part of the filter. The transformer to the right of the choke is the audio output transformer. To the left of the two tubes in front of the filter is the Relay. It appears that this might be a remote speaker on/off relay.

Operation - With no electrolytic filter capacitors and the majority of bypass capacitors being oil-filled, SN: 46 does function and seems to be a very stable and sensitive performer.

RAO Series

RAO & RAO-1 - National Co., Inc.

National began supplying the U.S. Navy with a special version of the NC-100XA as early as 1940. Actually, the contractor was General Electric Supply Company in Washington D.C. with National Company being the manufacturer of the equipment. The contract for the RAO-1 is NOs-70248 and the Contract Date is 11 Dec 1939. Designated as RAO, the first versions are somewhat similar to the standard NC-100XA receivers with a few exceptions. The civilian NC-100XA receiver used a field coil type speaker and push-pull audio output, neither of which the Navy would have wanted on their receivers. Consequently, the RAO and the RAO-1 have single-ended audio output along with an extra filter choke to take the place of the speaker field coil doubling as a choke. Early versions use National's 456kc IF but by the RAO-2 the IF was changed to 455kc. RAO receivers tune .54mc to 30mc in five tuning ranges. These early RAO receivers use the silver background dial along with the articulated dial pointer as "band-in-use" indicator. A single preselection stage was used in the RAO and the RAO-1. S-meters were included in both versions. A Noise Limiter was used in the RAO-1 but not in the RAO. The audio output tube used was a 6V6.

photo left: Although this photo is of the initial pre-WWII (June 1941) version of the Navy RBH receiver (National NC-156) it has the same general appearance that the initial RAO-1 receiver had.

The RAO-1 was a table version that added a Noise Limiter circuit and provided 5000Z audio for loudspeaker and 500Z for phones. A table loudspeaker was included that had a 5000Z matching transformer. Note that the matching loudspeaker is shown with a National Company cloisonné <NC> emblem installed (see artwork to the right.)

The US Navy RAO-7/9 manual along with the Navy "Catalogue of Electronic Equipment" are both very explicit in stating that the RAO and RAO-1 are not to be used onboard ship and are only suitable for "shore use" as they do not meet specifications. This undoubtedly was referring to the level of LO signal leakage coupling to the antenna, a problem that was corrected with the RAO-2 and later receivers by adding an extra stage of preselection (TRF amplifier stage) and additional shielding.

photo right: Artwork from the RAO-1 manual, contract NOs-70248

RAO-2 - National Co., Inc. - Long before WWII began, the Navy wanted minimal radiation from any of the shipboard superheterodyne receivers' Local Oscillator appearing on the antenna. This was primarily to allow the receivers to be used in the presence of other shipboard radio equipment without causing interference. Also, the receivers had to operate without interference from the other shipboard electronic equipment. Most Navy ships had multiple radio receivers and transmitters in close proximity. Most of the large Navy ships had multiple radio rooms and many ships were equipped with radar. It was imperative that all radio equipment could operate simultaneously without mutual interference. Additionally, as surveillance, signal intercept and shore communication operations increased at Naval facilities, with the many receivers operating simultaneously, mutual interference had to be eliminated for reliable and accurate reception.

There was also the highly publicized reason for "low radiation receivers." There was the remote possibility that an enemy could "Direction Find" (DF) the radio's position from receiving the LO radiation if it was allowed to couple to the antenna. It would be possible to also discern at what frequency the receiver was tuned. In the late-thirties, many of the Regenerative Medium Wave receivers used on commercial ships operating with autodyne detectors could easily be received at a distance of up to five miles or more - and this was from other commercial ships, not the enemy! Most of these types of regenerative receivers had been removed from ships prior to 1940. The Navy believed that if the Germans put their minds to it, they could probably receive and DF inadvertent superheterodyne LO leakage radiation on the ship's radio antenna from up to 100 miles away. Consequently, the Navy came up with a specification of <400pW as the maximum of LO energy that could be measured at the antenna input terminals of the receiver. This assured that the receiver operation would not cause interference and it would be impossible to receive the LO leakage as radiation from a ship antenna at any significant distance away from the ship.

Beginning with the second of the numbered suffixes, the RAO-2, National added an extra RF Amplifier stage for isolation of the antenna from the LO radiation. This was accomplished with an additional coil catacomb and tuning condenser housed in a bolt-on rear chassis and bolt-on cover. This added RF Amplifier upgrade dates from just before WWII with the upgrade possibly first installed on a U.S. Coast Guard receiver, the R-116. The R-116 contract was dated May 15, 1941 and, although there are a lot of similarities to the RAO-2, the mechanical design of the coil catacomb is somewhat different allowing for six tuning ranges by utilizing the National Co. NC-200 receiver catacomb. The circuit of the R116 has some significant differences from the RAO-2 the most obvious being that it is operated by a separate power supply. The RAO-2 simplified the coil catacomb gear drive, simplified circuit construction by eliminating the many chassis-mounted bypass capacitors and, of course, using the combination of NC-100 catacomb, the later NC-100 circuit (post -1939 with NL circuit) and using the NC-200 Crystal Filter. >>>

photo above: This snapshot of a couple of Navy shipboard radiomen at work shows the typical receivers used during WWII. Far left is an early RAO receiver on it's shock mount. Next to the RAO is the RBL TRF with regenerative detector longwave receiver. The larger receivers are the RAK and RAL, both TRF with regenerative detector receivers. Also, to the upper right is the LM frequency meter for accurately determining either transmitter frequency or received frequency. Below the order binders and just visible is a Scott SLR type shipboard entertainment receiver.

photo above: This is an enlargement of the background of an advertising photo that National used during WWII showing a RAO-2 equipped communications-relay station located in Africa.

>>> The second RF amplifier gain was set at unity-gain, so the overall receiver sensitivity is similar to the single RF stage NC-100A although with the extra coil set (TRF stage) the image ratio is improved and LO isolation from the antenna is increased. Antenna input shielding was also improved but the RAO-2 retained the exposed Antenna and Ground terminals (changed to a coaxial input with the RAO-3.) This additional Antenna Input/RF amplifier section (sometimes called "the back porch") increased the depth dimension by about four inches and increased the weight significantly, running the scale up to about 75 lbs for most of the RAO-2 and subsequent receivers (see photo of the underneath of the RAO-3 chassis in the next section below.) Tube total in RAO-2 (and later) is eleven. The contract for the RAO-2 was NXs-4681 and the date issued was April 30, 1942. The USN designation for the receiver was CNA-46187.

The RAO-2 borrowed several features that National had been using in their NC-200 receiver (introduced in October 1940.) As mentioned, the Crystal Filter circuit was the same that had been used in the civilian NC-200 receiver in that a stepped-switch Selectivity control provided five positions of increasingly narrow bandwidths and the variable Phasing control adjusted the "notch" position within the passband. Using this Crystal Filter required using 455kc as the IF but that made sense since 455kc had become an industry standard (more or less.) The LO tube was changed from a 6K7 to a 6J5 as in the NC-200.

Other RAO-2 changes were that the dial scale background was changed from silver to off-white (cream color) paint and the delicate articulated dial pointer was eliminated. Instead, a rotating dial gear-driven by the band change shaft indicated which tuning range is selected. RAO-2 and RAO-6 receivers had an ID from National of NC-120 on the control panel for the Crystal Filter (along with the National Co. identification and the "NC" diamond insignia.) The S-meter was from Marion Electric and the scale has the National diamond logo on the face. The chassis and most component housings are painted National gray on the RAO-2, only the IF, the BFO and the filter capacitor housings were matte aluminum finish. For reliability, no electrolytic capacitors were used in any of the RAO receiver versions. The power supply filters are 4uf oil-filled paper-dielectric types and 1uf paper-wax capacitors are used for filtering within the receiver circuitry.

The early RAO receivers (pre-RAO-7) are 17.5" wide, with an integral panel-cabinet-chassis construction that requires major disassembly if more than tube replacement, alignment or minor repair is needed. These early versions of the RAO used a cradle-type shock mount that is seldom found with the receiver today. Note in the "snapshot" radio room photo above-right and the photo to the left that the RAO receivers are mounted in their proper cradle-type mounting bases. Photo of the RAO-3 below also shows the cradle-type mounting base. The mounting base was identified as CNA-10125 when supplied with the RAO-2.

photo above:  Wells Gardner & Co.  RAO-3  SN: 1064 installed on CWQ-10125-A mounting base

Wells Gardner & Company (Chicago, IL) was the second contractor for the RAO series, building the RAO-3, 4 & 5. The RAO-3 and RAO-4 were built on contract NXss21446 dated Jan. 11, 1943. The RAO-3 Navy identification is CWQ-46187-A and the RAO-4 was CWQ-46187-B. The RAO-5 was identified as CWQ-46229 and was built on contract NXsr38492. The RAO-5 appearance is almost exactly like the RAO-4 but an accessory loudspeaker is listed on the data plate. The mounting base was identified as CWQ-10125-A for RAO-3, 4 & 5. It's the same number as the RAO-2 mounting base but with the Wells Gardner identification CWQ prefix and an "A" suffix.

All WG versions have the extra RF amplifier for double preselection and all versions have S-meters installed. Though WG used National Co. "designed" parts, where specified, many other components were strictly WG manufactured or purchased parts. The IF transformers, power transformer, chokes and the chassis all appear to be WG components (that is, built to spec or purchased by WG but differing in appearance when compared to the same National component) while the knobs, dial, tuning condenser and coil catacomb appear to be directly from National. However, close examination of these "National" parts will reveal slight differences. For example, the catacomb contact insulators used in National receivers were made of a material called R-39 but the WG insulators are molded bakelite. When a National "bar knob" is placed next to the Wells Gardner "bar knob" it's obvious that the WG knob is slightly smaller, more squared and the pointer is much narrower.

There was a myth that Wells Gardner RAO versions were not considered to be "heavy-duty military-type" construction by the Navy and that the Navy directed that WG RAOs could only be used at shore stations. But, the WG Navy manual specifically states the receiver is for shore stations OR for shipboard use since its LO leakage to the antenna is <400pW. Also, any examination of a WG RAO will reveal that its construction is just as robust as the National RAO-2 or RAO-6. The myth probably originated because nearly all RAOs receivers were used on shore for surveillance and communications. Certainly when examining WWII vintage photographs of radio set ups, the RAO receivers are seldom seen on ships but are very common at shore stations. Certainly, the lack of any RAO having medium wave Marine Band coverage might have been the reason for their limited shipboard use (the RAO was usually paired with the RBL for MW-LF-VLF coverage onboard ship.) For signal intercept, frequency surveillance and communications relay the RAO was ideal with wide frequency coverage per band, the ability to tune from the low end to the high end of a band quickly along with good dial accuracy and stability. Perhaps a more accurate version of the myth might be that all versions of RAO receivers were better suited for the types of duties found at Navy shore stations than they were as "shipboard" receivers during wartime.

The RAO-3 operated on 115vac only while the RAO-4 and 5 had a dual primary power transformer to allow operation on either 115vac or 230vac. The audio output impedance was specified as "total load of 600 ohms" referring to the possibility of multiple headsets being connected to the receiver. The RAO-5 included a loudspeaker with 600Z matching transformer (supplied by Jensen, CJS-49493.) Interestingly, the WG versions of the receiver had rivet mounted metal glides or  "feet" on the bottom covers that allowed using the receiver on a table top without the shock mount assembly. When "in the crate," the shock mount, the manual and a spares kit (an actual metal box of extra parts) were packed with the receiver.

Marion Electric S-meter for National RAO, RAO-1, RAO-2 and the early RAO-6 receivers.

Marion Electric S-meter for Wells Gardner RAO-3, 4 or 5

Also note that all paper-wax capacitors, other than all .01uf values and one .05uf and one .1uf, utilize a clamp to mechanically secure these physically larger capacitors to the chassis. All carbon composition resistors were supplied by Allen-Bradley and are in the JAN style. These resistors are known for the long-term stability and reliability. All tube sockets are isolantite types (ceramic tube sockets.) The RAO-3 is the first RAO version to use a concentric cable antenna connector rather than a thumb screw terminal block.

Surveillance, Intercept and Communications Relay for the RAO Receivers

The large device in the left foreground is a General Radio LR-1 precision heterodyne frequency meter. The LR-1 provided precise measurement of a transmitter's operating frequency or allowed the operators to set-up their receiver to a specific reception frequency accurately. The LR-1 was accurate to better than 100hz at 10mc if set up carefully.

By late-1943, it was becoming apparent to the USN that the primary use for the RAO receivers was either for enemy signal intercept or for frequency surveillance. The RAO receivers produced later in WWII were built specifically for use with a panoramic adaptor for signal frequency surveillance, beginning with the RAO-6.

photo left: Banks of RAO receivers used at a shore relay communications or intercept station. The receiver in the foreground shows the "back porch" addition of the 1st RF amp very well. Also, note the length of the USN coaxial plug installed in the antenna input receptacles. Since these receivers all have the coaxial antenna input and don't have the panadaptor inputs on the rear and are in the multi-piece cabinet, these are all Wells Gardner & Co. versions of the RAO receiver.

photo above: Using the Wells Gardner RAO receivers at a shore station for copy on the Underwood SII "mill" during WWII.

Apparently, at the beginning of the production run for RAO-6 receivers, the USN decided they needed to change the RAO-6 and optimize it for surveillance and specifically to function with a panoramic adaptor. The panoramic adaptor worked from the Mixer plate and allowed instant visual monitoring of wide segments of the spectrum which increased the ability of the radio operator to intercept enemy signals. The S-meter was considered superfluous since the panoramic adaptor also showed relative signal strength so the S-meter was to be eliminated.

The USN decision for a panoramic adaptor version RAO-6 should have required another different contract but it's uncertain whether that happened. It would also seem likely that a different Navy identification would have been used. The later panoramic adaptor RAO-6 receivers have a USN identification of CNA-46187-D, which is the same numerical designation as the RAO-2 but with a "D" suffix. One would assume that the S-meter version RAO-6 used suffix "C" and that suffixes "A" and "B" were for various contracts for the RAO-2 (but this is just an assumption.)

It should be noted that these "early" RAO-6 receivers would be very similar to the RAO-2 in features and if the data plates were missing it would be very easy to confuse a receiver's true identity. As listed in the Navy "Catalogue of Electronic Equipment," the initial (serial numbers 1 to 200) RAO-6 receivers were the S-meter versions which would make the early RAO-6 a rarity. The RAO-7 manual (NAVSHIPS 900,554) also mentions that the RAO-6 was equipped with a S-meter. The later, panoramic adaptor version RAO-6 is normally the receiver type that's encountered. Serial numbers 201 to 1200 (Navy Catalogue) were assigned to these later RAO-6 receivers. A different POWER SUPPLY B+ nomenclature plate is used on these later RAO-6 receivers and the data plate used on the top lid is smaller. The RAO-6 chassis and some components, such as chokes, power transformer cover, etc., are painted National gray.  

photo above: RAO-6 CNA-46187-D equipped for panoramic adaptor use. Note that "NC-120" is shown on the crystal filter panel. This RAO-6 belongs to N6MKC who supplied the photo.
 

RAO-7 and RAO-9 - National Co., Inc. - The RAO receiver underwent significant changes with the RAO-7 and RAO-9 (the designation RAO-8 was not used.) The RAO-7 used CNA-46233 for its Navy designation and the RAO-9 used CNA-46263. These receivers were of robust construction and had increased shielding to further allow their use with other shipboard equipment without mutual interference. The audio output Z was the standard 600 ohms and the design eliminated the S-meter in favor of a panoramic adaptor connection. The RAO-7 and RAO-9 simplified the maintenance of the receiver by designing the chassis so that it was easily removable from the cabinet (handles were added to the receiver front panel to assist removal.) The receiver front panel was changed to a heavy 8 gauge steel panel although the dial escutcheon remained about the same as earlier models. The cabinet itself was redesigned for better shielding and easier mechanical construction by making it a one piece unit. While the early RAOs used a separate mounting base, the new RAO-7 and RAO-9 cabinets mounted the shock feet directly to the bottom of the cabinet further easing construction and maintenance. The RAO-7 and RAO-9 receivers are physically larger than the earlier versions with full 19" rack width panels although the receivers are not specifically designed for rack mounting (since the cabinet provided necessary shielding.)
 

photo left:  National RAO-7 sn:10 from the first contract, NXsr-38306  Sept.22, 1943. Received by the Navy in September 30, 1944 but never put into service.  Chassis sn: is H720 which is National's actual production run serial number. The data plate serial number and the chassis serial number will never match on these receivers - they aren't even the same format. The chassis serial number was assigned before assembly was complete and is a letter-prefix and number suffix format. The data plate serial number was assigned when the receiver was used to fulfill the Navy contract and is strictly numerical. Other than the obvious major changes, note that the new RAO-7/9 also uses different style bar knobs that employ a skirt

Other minor changes were to the bar knobs that now had skirts added, some of the nomenclature panels were removed and replaced with individual circular control nomenclature metal plates. The data plates were moved from the top lid of the cabinet to the front panel of the receiver. The gear box was slightly changed to have an additional idler gear and internal stops.

Electronically, there is no difference between the RAO-7 and the RAO-9 with the exception that, in the RAO-9, the two RF amplifier inputs each have a RC filter installed between the grid and the coil selected. These RC filters were low-pass types to keep VHF radar from interfering with the receiver's operation onboard ship. 

Original receivers will have the Navy-style coaxial connector. Since these connectors are relatively difficult to find, many RAO receivers will have SO-239 coaxial connectors installed. If you still have the original Navy coaxial connector on your RAO, use a double female UHF coax adapter. The adapter may fit better if the threads are slightly filed until the adapter fits fully into the Navy connector and contacts the center pin. Then just connect the PL-259 up for the antenna connection. That preserves the rare, original Navy connector. Audio output is available either at the rear terminals or the front panel PHONES jack. Both are direct connections to the 600Z output transformer.

RAO-7 Serial Numbers and Contracts - There were two contracts for the RAO-7 (NXsr-38306-RAO-7 and NXsr-55614) and two contracts for the RAO-9 (NXsr-38306-RAO-9 and NXsr-85045.) A total of 1838 RAO-7 receivers were built and a total of 202 RAO-9 receivers were built. All contracts were issued between Sept.1943 and Dec.1944. Serial numbers on the data plates were assigned when the fully assembled receiver was used to fulfill a USN contract. The serial number on the chassis was assigned and stamped prior to production assembly of the receiver. These two serial numbers never match and shouldn't. They are different formats with the data plate SN being just numerical and the chassis SN being a letter prefix followed by numbers.

Data plate serial numbers appear to be non-sequential on these receivers. Serial numbers within the sequence may have been assigned to different accessories within the contract (e.g., the spares kit.) In fact, some data plate serial numbers are higher than the total quantity of receivers built for the contact. From this, one can infer that the numbers within the sequence must have been assigned to other items. To further the confusion, the two "J" production run RAO-7 receivers examined appear to have had their chassis serial number restamped (with the original stamping obliterated.) The build quantity of each production run is unknown but NAVSHIPS 900,356 indicates that 765 RAO-7s were on the first contract and 1073 RAO-7s were on the second contact. However, it appears that production runs aren't directly related to the contract quantities since it appears that National built somewhat ahead and then used completed receivers to fulfill the contracts.

More confusion can result in the possibility that the data plates found on a particular RAO receiver may not be original to that receiver. Of four RAO-7 receivers that have been examined closely, two positively have data plates that came from other receivers, another receiver's data plate originality is suspect and only one of the four (SN:10/H720) is positively the original data plate. The upshot of all of this is,...consider yourself lucky if your RAO-7 even has a data plate since many were swapped to different receivers during WWII for various reasons and many more were entirely removed post-WWII. Contemporary collector's have also swapped data plates from junk receivers to receivers being restored.

The stamped chassis serial number will provide some indication of the receiver's approximate build date. From reported serial numbers, it appears that production run prefixes start with "H" and end with "K" and this may help with possible dating with the early prefixes dating from late-1943 and the late prefixes being around early-1945.

photo left: RAO-7 SN: 10/H720. Note that the separate chassis for the 1st RF catacomb is still in use (note the junction "seam" just at the edge of the power input panel bracket.) The side panels provide rigidity for the chassis since it is now mounted in a complete cabinet rather than the sheet metal "pieces" that make up the earlier type cabinet. Other improvements are to the antenna input shielding and the addition of a panadaptor output - the SO-239 connector. Note the almost "sealed" RF amplifier box which requires removing the top cover to access the first RF amplifier tube. This receiver interior was never given a MFP treatment. Note that the RAO-7/9 chassis is cadmium plated.

Using any stock original RAO receiver paired with a transmitter will require using the B+ OFF position of the POWER SUPPLY switch for standby. As a station receiver the RAO will have very good sensitivity up to about 20mc or so (a directional "gain" antenna such as a yagi or quad should be used for higher frequencies,) excellent selectivity (use the crystal filter,) freedom from images, very good stability after a short warm up and good communications audio if a properly matched speaker is utilized. Since the dial readout is vague due to limited resolution, use a frequency meter (or RF signal generator) in the initial setup to exactly tune desired receive frequencies. The logging dial and scale were provided for accurate reset ability so write down the logging numbers and afterward use the receiver's logging dial and scale for accurate receive frequency set up. Best performance will be attained using a matched, resonant antenna. Antenna Input Z is approximately 70 ohms. Using a directional gain antenna such as a yagi will allow operation up to 30mc. I used my RAO-3 with a rotatable dipole as a Novice for 15M crystal-controlled CW operation in 1970. I was able to work as far away as VE3BAW in Ontario, Canada (on 21.212mc.) However, most vintage military radio nets operate in the 75M phone band nowadays where the RAO does a fine job with a typical wire dipole antenna although QRM will sometimes require using the Crystal Filter or "off frequency" tuning.

The RAO-2 and later versions do not use any electrolytic capacitors in the circuit. This has resulted in a typical, good condition, "as found" RAO receiver that will function fairly well on "all original parts." The paper-wax capacitors were the best that could be purchased at the time and the resistors were generally JAN-types that seem to be very stable as far as value drift. Most of the problems with the RAO receiver today are due to poor storage conditions, dings, dents and crunches that probably resulted from the weight of the receiver and trying to move it around in a storage environment. The cellulose-acetate dial cover will warp, darken and shrink in poor storage conditions and that is a very common problem but, luckily, it's very easy to fix. Careless or incompetent workmanship is also encountered fairly often and certainly "hamster" modifications also do show up from time to time. I think the overall size and weight of the RAO has greatly limited the amount of ham-induced problems,...it was just too big and exhibited way too much black wrinkle paint for a station receiver for most hams when it became available surplus in the 1950s. Excellent condition RAO receivers that have always been stored inside in a favorable environment and do operate should be checked over to be sure that nothing is getting hot during operation and that the voltages are about where they should be. At that point, the receiver should be aligned for best performance.

When I restored my RAO-3 with new rebuilt capacitors, I did notice an improvement in the audio but there were so many non-original parts in the circuit, it's difficult to tell whether the capacitor rebuild did much other than remove a "worry factor" about operating an electronic device with three quarters-of-a-century old paper dielectric capacitors. I don't intend to "re-cap" my all-original RAO-7 even though its components are just as old. It was stored inside at the Alameda Navy Air Station for years where even the outside climate doesn't change much. This type of consistent temperature storage environment has certainly helped to preserve the components and the cosmetics. Certainly poor storage in outside sheds or unheated garages will result in the most damage and these types of receivers will certainly benefit greatly from a thorough rebuild that includes new polyfilm caps to replace the paper-wax originals. On the 4uf oil-filled, paper dielectric filter capacitors,...check them for value and look for any seeping of the oil. If the value is correct and the seal is good, then these capacitors are very stable and will function correctly indefinitely. Be sure to check these capacitors at full B+ voltage and monitor the leakage current using an analog current meter. If you see any sudden meter needle movements that usually indicates arcing inside the capacitor. It doesn't happen too often but it should be checked to be sure the oil-filled nature of the dielectric is functioning correctly at voltage.  

RAO Accessories

Navy Department - Panoramic Radio Corp. - RCX Panoramic Adaptor SN:46 - Using a Panoramic Adaptor with the RAO-7 (455kc IF) makes for an interesting receiver set-up. Panoramic Adaptor coupling is thru a series 50K resistor inside the RAO-7 internal coaxial cable from the Mixer plate. There is a SO-239 connector on the rear of the RAO-7 and also a SO-239 on the rear of the RCX panoramic adaptor. All that's required is to connect the two SO-239 connectors together with a RG-58 cable that has PL-259 plugs on each end.

WWII panoramic adaptors aided surveillance by allowing the radio operator to see instantly if any unknown signals appeared within the slice of the spectrum he was monitoring. This eliminated the constant tuning that was necessary for receiver-only surveillance or for "guarding frequencies" using multiple receiver set-ups. The panoramic adaptor was primarily used for intercept of unknown signals. It wasn't necessarily used for signal analysis other than it was obvious to the radio op if the signal was CW or if it was somehow modulated. Surveillance and intercept was the panoramic adaptor's main role in WWII.

Shown in the photo to the right is Navy RAO-7 SN:10/H720 operating with Navy RCX Panoramic Adaptor SN:46. The RAO-7 is tuned to a weak SW BC station that is about 20kc down frequency from Radio Havana which was operating on 11.75mc in the 25M Shortwave Band. Radio Havana presents a strong AM signal as shown by the large peak at the right of center of the display. The total slice of spectrum showing is about 100kc or +/- 50kc each side of center. By tuning the RAO-7 up frequency 20kc, the large peak would move to the left on the CRT and, when centered, Radio Havana would then be within the RAO-7 IF passband and would be receivable. Scale is -50 -0+ +50 in kilocycles. Other graduated scales were available.

The RCX Gain is usually adjusted to just show a slight indication of noise at the bottom of the display with the receiver not tuned to a transmitting station. This will generally show any signal above the noise level. Extremely strong signals may require a slight reduction in the RCX Gain to show the peak of the signal. Also, receiver RF gain can be reduced if necessary. If the sweep is expanded, more details of individual signals can be observed. At +/- 50kc sweep only general characteristics can be observed. Modulation can be easily seen although its rapidly changing characteristics on the display doesn't photograph well. Comparative amplitudes can be easily seen.

The adjustments behind the "toilet seats" are for setting the display to be positioned correctly within the graduated scale. Intensity and Focus adjust the quality of the trace. Center Frequency allows other receivers with somewhat different IFs to be used. The RAO-7 IF is 455kc but the HRO receivers used 456kc and the Hammarlund Super-Pro receivers used 465kc. Any of those receivers are well within the adjustment range of the Center Frequency control (450kc to 475kc.) Sweep allows setting the approximate spectrum range of display from 0kc up to 200kc. The setting of ~ 100kc allows the -50-0-+50 scale installed to be easily used to measure where other signals are located in reference to the tuned signal.

CNA-10125 and CWQ-10125-A
Mounting Base

The National Co., Inc. RAO-2 and RAO-6 used mounting base CNA-10125. The Wells Gardner & Co. RAO-3, 4 and 5 used the same mounting base but with the designation of CWQ-10125-A. The receivers will have in each lower corner, both front and back, inserts that are mounted to the inside of the cabinet sheet metal and provide a threaded barrel for the four shock mount thumb screws to utilize for securing the mounting base to the receiver. The heavy duty "shock" feet have metal spacers that are suspended in rubber. When the RAO receiver was installed onto the desk where it was to be used and that installation was onboard a ship, then the mounting base was bolted to the desk. If installed in a shore station, whether the mounting base was bolted to the operating desk would depend on the situation. The photo left shows the top of the mount where the receiver is placed. The photo right shows the underneath of the mount and the shock feet (that were over sprayed with black paint - they should have a silver-color metal housing.)

RAO Loudspeaker

The RAO and the RAO-1 came with loudspeakers, a rack mount speaker panel in the case of the RAO and a table cabinet loudspeaker for the RAO-1. The RAO-2 and later receivers didn't specify a loudspeaker since the intent was for the operator to use 'phones for reception. The RAO-5 was the only later version RAO to include a loudspeaker in the list of supplied equipment. Actually, the entire loudspeaker was built by Jensen with a Navy identification of CJS-49493. It seems likely that the CJS-49493 was a "multipurpose loudspeaker" the was provided with other Navy equipment that required several nominal impedance selections and that it wasn't specifically made for the RAO-5 or any other RAO receiver.

The CJS-49493 loudspeaker was 6" in diameter, mounted in a black wrinkle finish box and the grille consisted of square wire mesh. Internally, there was a multi-tapped matching transformer that allowed selecting various nominal impedances from a low of 6Z up to a high of 20,000Z by way of a terminal board and wire connections. 600Z is one of the selectable impedances. Also included in the lower right corner of the front panel was a Volume control which was connected in series with the transformer secondary (lo-Z to the voice coil.) If the listening environment required the radio op to switch-over to phones, he could plug his phones into the RAO phone jack and then reduce the CJS-49493 Volume to the lowest setting (fully CCW) since plugging in phones didn't disconnect the loudspeaker. Typical of Navy loudspeakers, the connecting cable is a two-conductor shielded but un-jacketed cable. There is a grounding screw on the RAO receivers that is located near the SPKR terminals for connecting the loudspeaker shield to the receiver chassis-ground.

The RAO-7 and RAO-9 only list a loudspeaker as "equipment not supplied with receiver." The manual only shows a line drawing representation of a loudspeaker in one section and only the 600Z ohm impedance is mentioned.

The photo above-right shows the CJS-49493 Loudspeaker. The data plate is missing on this example but the Circuit Label inside the cabinet showing how to route the wires for the various impedances also shows the speaker's USN identification. The internal label is shown to the left.

Nearly all vintage photos showing RAO receiver set ups appear to have multiple operators using multiple receivers. All radio ops are using phones since multiple receiving stations precluded using loudspeakers. Today, any loudspeaker that has an input load of 600Z ohm will function correctly with any RAO-2, 3, 4, 5, 6, 7 or 9. The early RAO had a matching panel speaker with a 20K impedance matching transformer and RAO-1 used a 5000Z ohm output transformer and table cabinet speakers. Each would require a proper nominal impedance load to operate with the best efficiency.

Vacuum Tubes Used in the RAO and RBH Receivers

RAO-2 and subsequent versions:  RF1,2 = 6K7(2), Mixer = 6J7, LO = 6J5, IF = 6K7 (2), Det/NL = 6C8, BFO = 6J7, 1st AF, AVC = 6F8, AF Output = 6K6, Rect = 5Z3
Note: Audio Output tube was a 6V6 on initial RAO version

RBH, RBH-1 and subsequent versions:   RF = 6K7, Mixer = 6J7, LO = 6J7, IF = 6K7(2), Det/NL = 6C8, 1st AF/AVC = 6F8, BFO = 6J7, AF Output = 6V6, Rect = 5Z3
Note: Audio Output tube may have been 6K6 on CNA-46188 versions, also both RF tubes are 6K7 on CNA-46188 versions

Serial Number Log - Collected Serial Numbers for RAO Series and other Military Catacomb Coil Box Receivers

These serial numbers have been provided by other collectors that have responded to the "National Moving Coil Series of Receivers - NC-100 Series" article, specifically for the Serial Number Log. If you would like to send your RAO, RBH, NC-100ASD or R-116 serial numbers in, I'll list them here, in this log. Send your serial numbers to: RAO, RBH, NC-100ASD, R-116 SERIAL NUMBERS

RAO, RAO-1 Initial Models - National Company -

RAO-2 Series - National Company Build (RAO-2, 6, 7 or 9) -  10/H720 (RAO-7), nodp/J444 (RAO-7), 1642/J944 (RAO-7), nodp/K155 (RAO-7), 843 (RAO-2),

RAO-3 Series - Wells Gardner & Co. Build (RAO-3, 4 or 5) -  1064/na (RAO-3), 1412/na (RAO-3),

RBH Initial Model - 840,

RBH-1 Series -

NC-100ASD - Signal Corps - 194, 402, 426, 931, 948, A49**, A578

R-116 - U.S.C.G. - 46, 103,

xxx/xxxx = this format shows the data plate serial number first and the chassis stamped serial number second

nodp = no data plate,...serial number from chassis stamping

(xxx) = specific model receiver that sn assigned to

na = not available,...either not provided or not accessible or possibly not used

** = confirmed sn, although sn is in an unexpected format. A possibility is that "A" indicated "1000" and "A49" would have been receiver number 1000 + 49 or 1049. A578 would be "1578" - maybe,...just a hypothesis.

Restoring and Rebuilding RAO Receivers

How I Found the Wells Gardner & Co. RAO-3
A Brief Retelling of "Rios Radio Revisited" *

*"Rios Radio Revisited" was an article I wrote for "Antique Radio Classified" that was published in the October 1990 issue. It combined a "found radio story" about the RAO-3 with a bio of Phil Rios, an old mentor of mine.

The RAO-3 in 1990. This B&W photo for the "Rios Radio Revisited" article wasn't used.

Phil had been in the Radio-TV repair business since 1946 and I had worked for him during the summers when I was in high school. I knew he would have the tubes that were missing and some suggestions on restoration. Of course, in 1969, restoring a WWII relic was more of "get it running" than anything else. With the rocks removed, I saw that about half of the tubes were still present and we needed the easy ones, the 6K7, 6J7, etc. After "tubing up" the RAO, we found a power cord that sort of fit and a speaker (that was the wrong impedance - but who knew that then.) The RAO came to life and played pretty well with just a connection to the test TV antenna at the shop. Phil said, "Sounds pretty good. Run it for a week or so and then do an alignment and it should be a good set."

It wasn't too long before the audio output transformer opened up. I had taken the receiver down to Rios' shop and had the bottom cover off. After poking around for a while with no progress, Phil must have felt sorry for my inexperience and started issuing orders. "Measure the 6K6 pin 3. Okay, now measure pin 4. Okay, you have an open audio output transformer." Phil knew all the connections in his head from years of experience. I went over to Schirmer's Electronics, the local parts house, and bought a small universal audio output transformer. Who knew that the RAO was supposed to have a 600 ohm audio output impedance? Well, I installed the new transformer inside the old transformer's housing and everything looked original but of course the output impedance was now 8 ohms. The RAO came back to life and was ready for use as my new ham receiver. How I made any contacts on 15M CW I'll never know, but I did. As far away as VE3BAW (for one thing, I used a 15M rotatable dipole for the antenna.) Exciting stuff then. One thing to consider,...in 1969 this RAO-3 was only a 26 year old receiver (the paper caps have aged over half-a-century since I used this receiver as a Novice.) I doubt the performance today on all original parts (in 2021) is anything like it was in 1969. I later replaced the RAO with a Collins 75A-4 which was a vast difference in performance to a fairly new ham. I didn't get rid of the RAO though. I've always kept it in storage through several moves.

Plexiglass Dial Cover - Who knew this cover was supposed to be plexiglass? Not me, obviously, since I had installed a pane of glass. I guess that assumption was based on seeing an old NC-100ASD at the neighboring ham's shack when I was a kid (that was Dennis Anttila, WA6LHL, another mentor of mine.) I remember that his receiver's dial cover was cracked, so it had to be glass (glass IS correct for the ASD.) Like most RAOs, the plexiglass on this one was severely scratched up and discolored, so glass went in. Thirty years later, after realizing for some time that the glass was incorrect for the RAO, during this rebuild I replaced the glass with a correct plexiglass dial cover. NOTE: The new material used is Lexan, which is available in the correct .090" thickness. Lexan is highly resistant to cracking or discoloring. Original RAO dial covers were cellulose acetate, a very unstable plastic that was photosensitive resulting in darkening to a semi-transparent brown color. It could also easily shrink in size, warp and could become somewhat opaque if exposed to a lot of moisture.

Interesting Painting Problems - At first I thought I was going to be able to just spray over the original wrinkle finish with new wrinkle finish paint and the cabinet would look okay. Luckily, I tested the idea on the dial escutcheon first. Wells Gardner used some sort of dark olive drab primer on the sheet metal that reacted with the new paint by "lifting" - almost like paint stripper was being used. So, with that idea eliminated, I had to strip each of the cabinet sheet metal pieces first. This was probably better anyway, since the paint job will be higher quality.

The RAO-3 cabinet pieces were prepped by light sanding with 400 grit Al-Ox paper and washing with lacquer thinner. You have to paint one side at a time because the wrinkle paint has to be hot to "wrinkle." I use heat lamps along with a heat gun to assure that the wrinkle is complete. Since the lamps are in a fixed position, only one side at a time can be painted and then put under the lamps. Apply at least three heavy coats of wrinkle paint with a time spacing of about 1 or 2 minutes between each coat. Then place that painted side under the lamps and standby with the heat gun. Wrinkling will take place in about 5 to 10 minutes. Use the heat gun on the corners and edges where the lights don't provide enough heat. Don't use too much heat from the heat gun or you'll "gloss" the wrinkle and it won't match. You don't need a lot of heat, just moderate heat to assure the you get full wrinkling on the entire panel-surface. If you're careful painting each side to not "over-shoot" the edge you'll find that each side will blend and when finished the cabinet looks like it was all painted at one time. Be sure to allow a little time for curing before assembly. Overnight is okay but a few days is better and a month is about how long the paint needs to fully harden.

Shown above is the main cabinet piece after black wrinkle finish was applied.   NOTE - this application is using Krylon Black Wrinkle Finish paint, not the VHT paint I now use. At the time I painted the RAO in 2006, only Krylon wrinkle finish paint was available. When the Krylon is compared to true "black" paint, one will see that the Krylon is sort of a charcoal black, not deep, jet black.

The original RAO-3 shock mount turned up when a local collector-ham bought the shock mount thinking it would work for his RBL receiver but the RBL cabinet has less depth so the shock mount thumbscrews couldn't be tightened at all. Since it wouldn't work for the RBL, the shock mount was given to me for the RAO-3. I believe this was around 2015.

The filter capacitors are not electrolytics - there are no electrolytic capacitors used in the RAO receivers. The filter caps are oil-filled paper dielectric capacitors which is why most of the RAO receivers will usually work somewhat with mostly original parts. Replacement of the paper-wax capacitors would be recommended for best and safest performance, however I didn't do that since I wasn't planning on using the RAO-3 as a station receiver. Maybe that will change in 2021,...and it did.

Further Inspection of the "Old RAO-3"

1. All tubes tested good except 6J5 LO tested odd. Decreasing gm as test button held down which usually indicates a tube at the end of its life or possibly a gassy tube. Might have caused the decreasing sensitivity (it did.)
2. Measured all resistors in circuit. All are JAN Allen-Bradley resistors. Most measured within 10%. One measured higher than marked value which indicates actual value drift. Several measured lower but this can be parallel resistance within the circuit. JAN Allen-Bradley carbon resistors are just about the most reliable and stable vintage types of carbon resistors. Only long term heat caused by leaky bypass capacitors usually will affect the value. Later testing the suspect resistor that appeared to be 10K but measured 70K - a schematic check indicated the resistor IS a 70K and the violet color band had faded to a brownish color. So, all JAN A-B resistors were okay. 
3. Three of the original mica capacitors had been replaced with dipped silver mica capacitors. I'm sure I didn't do this, so these replacement parts are from before 1969.
4. Three of the original paper capacitors were replacements. All three are different brand paper wax caps and the soldering job indicates these were probably the result of a pre-1969 junk box repair work.
5. AF Gain pot is an A-B replacement pot. It appears to be a linear taper type since most of the gain adjustment is from 0 to 3 and above about 5 there isn't any increase in audio gain. The A-B pot measures 275K at 50% and the manual indicates that 25K at 50% (which is an audio taper) is the original spec for the pot. Overall output level during this test was undoubtedly affected by the 600Z loudspeaker. Taper not affected, still needs audio taper.
6. Paper capacitor count is .01uf-600vdc = 9, .1uf-400vdc = 13, .05uf-600vdc = 2, .1uf-600vdc = 4, 1.0uf-200vdc = 4, that's a total of 32 capacitors. OEM was John E. Fast & Co. in Chicago. Capacitor shell only has CBV, which was John E. Fast & Co. Navy contractor identification along with a part number, value and working voltage. Also, a letter-type date code.
7. Under the 1st RF Amplifier (the metal box mounted to the back wall of the cabinet) are three paper-wax capacitors. All are .1uf 400vdc. These capacitors are next to impossible to see unless maybe a dental mirror is used. They are also impossible to remove and replace with the RF Amp box mounted to the back wall of the cabinet. Also, it's next to impossible to dismount and extract the RF box unless the 3rd IF transformer is also removed from the chassis. To extract the box also requires unsoldering three wires. The almost fully hidden capacitors and the difficulty in doing a replacement has probably resulted in many "recapped" RAO receivers having "originals" of these three caps still in place.

Initial Rework Necessary on the "Old RAO-3" 1. Quite a bit of surface rust in one area inside the "back porch" chassis. Wire brushed to clean and then applied 10W machine oil to protect. After a few days the oil dried up, then the area was cleaned with lacquer thinner and painted to protect the surface. I'm sure the rust has been there all the time I've had the receiver. The receiver has been in Nevada since 1973 and has always been indoors since 1969, so the moisture that caused the rust was from before I owned the receiver. Inside the "back porch" there was some rust in the joints which was cleaned and repainted with black nitrocellulose lacquer. 2. The seven pin "battery input" socket had been rewired and doesn't agree with the schematic. I thought initially that someone had rewired it for remote standby. Five out of seven wires were moved to different pins and the A+ wire was connected to chassis externally (hamster job with the wire wrapped under a screw) instead of through the socket with ground lug connection to chassis. Unfortunately the wire insulation is white with barely perceptible colored tracers. Also, at least one wire isn't original. Found a "cut and splice" taped with masking tape hidden by the rectifier wiring harness that was part of the one non-original wire. Since the CT or the B+ switch weren't part of the changes, it's apparent that the changes weren't for remote standby. All wires had to be traced using a DMM (in ohms) to confirm what the wires were and where they should be connected. This required unsoldering all of the wires from the socket, tracing the wires and then applying a masking tape numbered identification. After all of the wires were identified then the socket pins were cleaned and the wires soldered to the proper pins. The mating plug had to also be reconfigured to have the proper connections within. The two jumpers in the plug are for AC input and for tube heaters. While a remote standby function might be convenient, none of the RAO receivers originally had it and almost all RAO users know that the front panel standby has to be used when the receiver is paired with a transmitter. Also, for the future, it makes it much easier to work on a receiver if its wiring agrees with the manual. This socket was originally for emergency battery operation of the receiver which was a Navy requirement at the time of the contract. I believe that the mod-changes were to allow remotely applying AC power to the receiver (maybe an external timer to turn the receiver on or off at specific times.)   3. Replaced the SO-239 flanged UHF receptacle that had been installed in place of the original Navy coax receptacle. Luckily, no holes drilled (flange was on the inside and retaining nut on the exterior.) Rebuilt the antenna input to match drawings using an original Navy coax receptacle. Almost all RAO receivers have had their original Navy coax receptacle replaced with a UHF receptacle. This really wasn't necessary and is just another "hamster" modification. On some Navy coax receptacle the internal pin is exactly like a male banana pin and a female banana receptacle will fit inside the Navy receptacle allowing direct connection for the coax center conductor and the coax shield can be connected to the screw ground just above the receptacle. Other Navy coax receptacles are slightly larger and a double female UHF adapter can be used. The threads have to be filed down slightly on the side that will fit into the Navy receptacle. Once the adapter fits, the center conductor of the adapter is the correct size to fit the Navy receptacle pin. Just push in the adapter and then screw the coax PL-259 to the other end of the adapter and the antenna is connected without ruining the original (and hard to find) Navy coax receptacle. There are two different types, so you have to check which one you have. I had to do the banana plug receptacle type of connection, (receptacle change was with the RAO-7/9 introduction although it's possible that some RAO-6 receivers also may have the later receptacle installed.) Transplanting Parts - After removing all of the cabinet pieces from the old RAO-3, I noticed that the front of the BFO can and the front of the Crystal Filter housing were moderately bent as if the receiver was hit in in front near the Crystal Filter and BFO area. I do remember that when I found the receiver the Crystal Filter phasing control shaft was broken off and the original paint had two significant scratches down to the bare metal in that area. It might be an easy transplant to R&R the BFO can and the Crystal Filter assembly from the parts set into the old RAO. My old repair of the phasing shaft was slightly off-center and although it worked it did result in an obvious "off-center" rotation of the Phasing knob.

photo above: The "parts set" data plates shown now installed on my old RAO-3 using the original hardware from the "parts set." These original data plates are the celluloid plastic type of data plate rather than the etched metal tags.

Jan 17, 2021 - R&R the power transformer cover. Removed the original four adapters for the shock mount thumbscrews. Five bar knobs and the tuning knob are present on the parts set and they are original WG knobs. Will have to soak the set screws with penetrating oil as they are really stuck. R&R the 1st RF amplifier tube cover. Removed the replacement audio output transformer from the old RAO-3 and removed the original audio output transformer from the parts set. Installed the original audio output transformer in the old RAO. Installed a reconditioned Centralab 500K audio taper pot for the AF Gain control and installed. Jan 18 - Worked with stuck set screws on the six original knobs on the parts set. Penetrating oil and heat worked on all but one set screw that unfortunately had the hex rounded so it was impossible to remove with the proper tools. A reverse flute drill bit might be able to break the set screw loose to save this original bar knob. The 2006 restoration replaced all but two WG bar knobs with National bar knobs. Fortunately, the tuning knob and the band change knob are original WG knobs. At any rate, if I can salvage all the WG knobs on the parts set, I'll only be short one WG bar knob with all other knobs on the receiver being the correct WG types. Jan 19 - Removed the three oil-filled filter capacitors from the parts set. All three tested at 4.6uf with cap checker. Removed the two dial lamp harnesses from the parts set. Jan 20 - The set screw from the last bar knob wasn't going to come out so I used a Dremel tool with cut-off disk to cut the shaft of the NL control. With the knob removed now the brass shaft can be drilled out easily and the knob salvaged. With the last knob removed the entire front part of the cabinet was dismounted. This allowed easy access to the chassis components I needed to dismount. Removed Crystal Filter assembly. Removed BFO coil-capacitor assembly. Removed 2nd and 3rd IF transformers.  Jan 21- I removed the side cover on the soon-to-be transplanted Crystal Filter for a close inspection and to clean the Selectivity switch with De-Oxit. No problems were seen. Transplanted the Crystal Filter assembly into the old RAO-3. From the wiring wraps, it looked like the old Crystal Filter had been removed at some time in the past, hmmm. Also, the front bend was causing a misalignment of the Selectivity switch shaft. The replacement assembly sets "square" on the chassis and the two shafts are straight. When I was removing the old BFO can, I noticed that the BFO tube socket was badly broken. From the top it looked like just a hairline crack but from underneath a large section of ceramic was missing and one tube pin receptacle was hanging loose. I guess it has been that way all the time I've had the receiver. At any rate, I replaced the broken socket with the BFO tube socket from the parts set. Then transplanted the BFO can. The bend in the old BFO can was severe and looked like it stressed the variable condenser although it did function okay.

photo above: The RAO "parts set" with most of the major parts removed for reinstallation into the "old RAO-3." Parts that still need extraction are the RF Amp box and the 6C8 shield. The power transformer cover shown is the non-original from the "old RAO-3."

Jan 22 - Installed 2nd IF transformer. Removed 3rd IF transformer from old RAO-3.   IMPORTANT NOTE: When cleaning the chassis to install the replacement 3rd IF, I noticed that under the 1st RF Amplifier tube socket there were some paper-wax capacitors installed. These are impossible to see unless the 3rd IF transformer is removed (maybe a dental mirror might help to see them,...if you know they're there.) These three capacitors can't be replaced unless the entire 1st RF Amp assembly is first dismounted and removed from the receiver. There are three wires that need to be unsoldered to remove the entire 1st RF Amp assembly (besides the six screws that mount the assembly to the back vertical cabinet wall.) Additionally, the 3rd IF transformer has to be dismounted to allow the RF Amp box to be tilted over so the plate connection can be unsoldered. Even with the 3rd IF transformer removed the space is very restricted. If one knows about the RF box caps and is also planning on removing the IF transformers then, with the 2nd and 3rd IF transformers extracted, the space to remove the RF box is much larger making the job easier. Once the RF Amp box is dismounted, then access to the capacitors is not restricted. I had to rebuild the three paper-wax caps in order to be able to then remount the 1st RF amp assembly and then the 3rd IF transformer could be reinstalled. Photo to the right shows the bottom of the RF Amp box with the rebuilt original paper-wax capacitors (new polyfilms inside) installed. Brief Capacitor Rebuilding Description - Restuffing caps consists of first using a hand-held heat gun to melt the wax in order to remove the old capacitor from the shell. The shell is wiped (while still hot) with a paper towel to remove the excess old wax which results in a clean empty original capacitor shell. Next a polyfilm capacitor is "sized" by wrapping it with masking tape until it fits into the shell without falling out. Then the ends are filled with hot-melt glue leaving about .125" to .250" gap at the ends. The ends are then filled with melted bee's wax. When the end cap wax has set-up, then the entire exterior of the capacitor shell is coated with melted bee's wax. When completed, it's very difficult to tell the rebuilt from an original. Polyfilm cap orientation shouldn't matter but I always install the caps with the left lead as the "outside foil" when holding the capacitor in the right hand and the value writing is viewed as "rightside-up." Jan 23 - Cleaned the tray area under tuning condenser with denatured alcohol. This area always collects all of the debris that makes it into the receiver cabinet over the years. Lots of greasy dust type dirt that cleaned easily with the denatured alcohol using small long handle paint brushes and long handle Q-tips. Detailed cleaning on the tuning condenser was also done. Cleaned the chassis ground connections and then installed the three oil-filled paper-dielectric filter capacitors. The three oil-filled caps were apparently removed sometime in the past (before 1969) and not reinstalled correctly. The insulating washer was next to the capacitor body and the grounding washer was next to the chassis. The grounding washer should be next to the cap body to connect it to chassis ground (not an electrical connection but for shielding) then the fiber washer is next to the chassis. Not a serious mistake but still not correct. I'm beginning to discover that my old RAO-3 had been worked on extensively and not by any Navy depot. While the rework is generally okay, it wasn't at a professional level that would indicate depot rework.

photo above: The 1st RF Amp box showing the three rebuilt capacitors installed on the underside of the assembly.

Jan 24 - Started rebuilding the paper-wax capacitors. I'm removing a small quantity of original capacitors from the "parts set" and then rebuilding those. Once that quantity is finished I then remove the same value caps from the old RAO and then install the rebuilt ones. I rebuilt the six paper-wax caps on the power supply side of the chassis. These along with the three under the RF box total nine caps rebuilt and installed so far. Both the "parts set" RAO and the old RAO have a few non-original capacitors so a few of the same value caps from the old RAO will need to be rebuilt and then reinstalled into the positions that had non-original maker caps. The end-result will be that all paper-wax capacitors will appear to be original CBV types but will have modern polyfilm capacitors inside.

Jan 25 - Rebuilt the two 1uf 200vdc caps, the two .05uf 400vdc caps and the three 0.1uf 600vdc caps. Then installed them into the old RAO-3. 16 caps rebuilt so far.  Jan 26 - Rebuilt eight .01uf 600vdc caps. Jan 27 - Installed all of the .01uf 600vdc rebuilt caps. 24 caps rebuilt and installed. I removed the remaining .1uf 400vdc caps from the "parts set" with the exception of C189 that has a unique clamp mount. For C189 I used the original from the old RAO-3. These eight caps were rebuilt and then installed bringing the complete total of rebuilt paper-wax caps to 32.   photo left: Under the IF/AF side of the chassis showing how the rebuilt paper-wax capacitors are difficult to tell from the original CBV paper-wax capacitors. Note that only the .01uf paper-wax capacitors and one .05uf capacitor and one .1uf capacitor don't have clamps to secure them to the chassis. photo right: Under the power supply side of the chassis showing the rebuilt CBV paper-wax capacitors located there.

Jan 28 - Replaced the three newer style dipped mica capacitors with original style, tested vintage mica caps (this was for under-chassis appearance only.) Some of the original mica capacitors used in both WG RAO-3 receivers seemed to have defective leads. The lead wire seems to be very brittle and easily breaks when trying to extract the capacitor. One of the original receiver problems back in 1969 was due to a broken lead on a mica capacitor. It's probably why the three dipped mica caps were installed. My thought is that if an original mica cap can be extracted from the "parts set" without damage then the leads are probably not defective. About 50% of the mica caps in the WG RAO-3 receivers seem to be affected. Part of the "test" besides measuring the value was to determine that the lead wasn't brittle.

Replaced the grid leads to the 6C8G and the 6F8G tubes. The "parts set" wires were in better condition than those in the old RAO. I was going to replace the S-meter lamp wiring because the two wires were "friction taped." But, the wires are within a laced harness and removal of the wires would probably break the laces. Instead, I replaced the friction tape with original black spaghetti tubing which looks more convincing as original. The wire from the S-meter switch to the meter also had friction tape on it so I used the good condition wire from the "parts set" as a replacement. Grid connection wire to the 2nd RF amp had friction tape so repaired with black spaghetti tubing. Installed the NOS S-meter and installed original lugs for the meter connections, also used original meter stud nuts and washers. Installed the original type of S-meter lamp socket. I had a screw base #40 lamp installed, should have been bayonet base #47. The rubber grommets on the lamp socket are to prevent pushing the lamp into the S-meter too far which can overheat the meter scale and discolor it. 

I added two fiber washers between the back of the logging dial hub and the front of the gearbox hub to take up some clearance that had been allowing the tuning shaft to have about .090" thrust end play. Mechanically set the logging dial to be accurate. 

Functional Test - Installed the tubes and cleaned the pins with brushed on DeOxit and then plugged in and out a few times. Connected 600Z ohm loudspeaker, the AC power cord and a test antenna. Power applied with variac with no problems. Full scale S-meter indication in AVC and no signal (AVC at max negative bias, apparently.) Loud signals in MVC, tested three bands with loud ambient noise levels, tested Crystal Filter OK, tested BFO OK, tested Tone OK, tested NL OK, RF and AF gain OK. Receiver is almost fully functional with just an AVC problem. When the AVC problem is repaired, then the receiver will be reassembled.

Jan 29 -  Even though the coupling capacitor from the detector/3rd IF transformer to the grid input of the AVC amplifier tube (a 50pf mica) seemed to measure okay on the cap meter, it apparently was open. Installed another vintage 50pf mica and that got the AVC functioning. Synchronized the band indicator to the catacomb position - it has to mesh with the pinion gear on the correct "tooth" of the drive gear in order for the correct band indicator to be in the 12 o'clock position when that band is engaged at the catacomb pin receiving contact blocks. Installed the combination front panel-side panels part of the cabinet. Installed the lid, the band change bezel, the dial escutcheon and the "back porch" cover. All original screws and lock washers used in all of the mountings. Most of the black wrinkle finish truss-head screws needed cleaning and touched-up. Installed the four threaded adapters that are for mounting the receiver to the shock mount. Lightly lubed (with grease) the two round rods that the two catacombs ride on.

Jan 31 - Operated the RAO-3 yesterday and this morning to allow any unusual issues to turn up but the receiver performed quite well. I was thinking about not doing an alignment because the sensitivity was okay and the tracking was very good. Luckily, I did do a full IF/RF alignment in the afternoon. The IF was slightly improved with the alignment. Tracking also was very close but Band C did need the lower end slightly touched-up to be accurate on the 40M band. Audio is very good although it is "communications grade" but the bass response isn't lacking, it just isn't emphasized. The problem of the slight frequency change in the BFO as the RF Gain is adjusted in MVC is now only encountered on Band A and even there it's just barely noticeable compared to what it was prior to the rebuild.

RAO-3 Performance - There's no doubt that after 50+ years of ownership,...it was about time that I rebuilt this receiver correctly. As I mentioned somewhere earlier, this RAO-3 was only a 26 year old piece of equipment when I found it in 1969. I had always run it on its mostly original parts that were now over 75 years old. So,...was improvement noticeable after all of the rework was finished? Absolutely! Here's what I've observed as definitely better performance: 1. Audio response - certainly replacing the "universal" audio output transformer with a good original RAO-3 audio output transformer was the major reason for the improved audio. A second improvement must have also been the replacement of the incorrect "linear taper" AF Gain pot with an "audio taper" pot that now results in a fully adjustable audio gain. Of course, slightly leaky capacitors can also skew bias voltages and affect low frequency coupling, so the capacitor rebuild also added to the much improved audio. Additionally, I had been running a beam power pentode 6V6 as the audio output tube and the correct tube is a pentode 6K6. NOTE: I didn't change any values of C, the values I used are original to the receiver design. 2. Selectivity - this improvement is due to the replacement of the damaged Crystal Filter assembly. The old unit had a severe "crunch" in the front that was putting stress on the Selectivity switch and the Phasing condenser with the misaligned wooden shaft prevented smooth adjustment. Accurate alignment of the replacement Crystal Filter also must have helped. The Crystal Filter is now fully operational and a valuable tool to combat QRM. 3. Sensitivity - when listening to 40M SSB nets the RF Gain is now at 4 or 5 with the AF Gain at 8. 20M SSB stations are normally between 5 and 7 for the RF Gain. Trenton Military on 15.065mc is easy copy. TAH (Marker Beacon, Istanbul, Turkey) copied on 8.35mc. All of the marker beacons from Seoul, Korea were easy copy at 12.9 mc. Nice improvement. 4. Dial Accuracy - this was always pretty good but is even better now with the recent RF tracking touch-ups. The best compliment for a receiver rebuild is to actually be able to enjoy using that receiver in a vintage military station while it successfully and repeatedly provides reliable communications with fine articulate audio,...just as seen in the photo to the right. This setup has been in use weekly use since early February 2021.

photo above: Left to right - WWII Underwood SII "Mill" typewriter, Navy-Collins T-47/ART-13 transmitter on top of the vacuum tube power supply (homebrew,) Navy RAO-3 receiver with a National MSC-10 speaker cabinet with 10" PM speaker with a 600Z matching transformer, Shure 102C carbon mike, a Hi-Mound HK-710 handkey and, on top of the ART-13,...a  Vibroplex Original "bug."

UPDATES: Dec. 16, 2021 - I'd been operating the RAO-3 directly on the 124 volt AC line which will slightly increase the tube filament voltage and also increase the B+ somewhat. The RAO was designed for 115vac input and the easiest way to adjust the AC line is to use a "bucking transformer." I looked through the filament transformer box and found a 6.3vac 6A transformer made by Triad. I wired it up to "buck" the voltage and the resulting AC input to the RAO was now 116vac. I just use power cable for wiring with an AC plug and an AC cable receptacle. All connections to the transformer leads are soldered and insulated then tie-wrapped to the mounting flanges for security. The bucking transformer just sets behind the RAO and is connected to the power strip that powers up the station. With that way of connecting the bucking transformer, it's only powered up when the strip is turned on and that only happens when the station is in use. No real obvious difference in RAO operation but the filament voltage is now 6.3vac and the B+ is around +200vdc. Oct. 11, 2022 - I've been listening to the RAO-3 lately and I'm beginning to wonder about the AVC. When switched in, the AVC throttles-back the RF/IF gain to the point where it seems that the AVC bias is dependent on the received noise rather than the signal. There doesn't seem to be very much change in the AVC bias in relation to the signal level. Although the S-meter appears to be working, on higher frequencies tuning off signal and back on signal doesn't really change the S-meter indication that might be reading S-6 or so. Performance in CW is great with good sensitivity easily up to 20mc. Another strange problem is the Noise Limiter that appears to be working in reverse with 9 or 10 seeming to be no limiting and max limiting at 0. It's difficult to find in the manual but the wiring diagram reference appears to indicate that maximum negative bias should be at 0 and that should be "no limiting." When I put the RAO-3 back on the workbench, I'll do some more investigating into these two very subtle problems. More details at that time,...

Return to Original, Mechanical Restoration of RAO-7  SN: J444 

 August - September 2018

Attention to detail is very important in any restoration. Correct mounting hardware and period finishes along with careful workmanship will result in an original appearance to any restoration. Without having a totally complete and original RAO-7 (SN:10/H720) as an easily accessible reference as to "what's correct?" I don't think I'd have been able to impart as much exacting detail into the mechanics of this restoration. Maybe this write-up might be "too much detail" on the mechanical aspects of the restoration but, if you're interested in minutia and appreciate the excruciating OCD-aspect of "attention to detail," then all of the work that was necessary in this restoration is covered in the sections that follow.

Two restoration attempts had been made by two different former owners over a period of a couple of decades. The former owners never seriously got into the electronics so the receiver chassis was all original. Having never been molested, the receiver actually functioned fairly well. Chassis serial number was J444.

The serious problems that kept J444 from being an excellent example of the RAO family were all mechanical and cosmetic in nature. The last owner of the receiver had considered the original front panel "not restorable" so he made a completely new front panel out of  8 gauge steel and painted it black wrinkle finish. Since this was a "brand new replica" front panel, none of the hardware needed to be of the original size or type. For example, the dial nomenclature plates were mounted with screws and nuts instead of drive screws or rivets. It ended up that every piece that could be mounted to the replica front panel was mounted with a non-original size or types of hardware. Although at a distance the RAO-7 SN: J444 with its replica front panel looked good, close-up the non-originality was very apparent.

I was extremely lucky that this last owner had kept J444's original front panel and that it was included with the receiver. The original front panel had been modified somewhat and this was why the last owner considered it "not restorable" and made the replica front panel. However, I thought the original front panel only had minor damage and with a little All-Aluminum Body Filler and wrinkle finish "touch up" it could be brought back to an excellent appearance and returned to its original receiver.

photo above: RAO-7 SN: J444 - after its mechanical and cosmetic return to original. This receiver was destined to go to the South Pacific so it was given a heavy MFP treatment. Besides the chassis, the front panel nomenclature plates and knobs were also MFP coated. This accounts for the yellowish color of the tags.

I used All-Aluminum Body Filler (ABF) to patch all of the holes. This was then sanded, primed and painted with black wrinkle finish (BWF.) I hand painted the ABF patches with BWF and then used a heat-gun to actuate the wrinkle. After this BWF paint had set for a few days, the entire panel was wiped down with thinned black nitrocellulose lacquer to even-out the color and bring out the luster. All of the patched holes except one were covered by data plates when the receiver was fully assembled. 

The escutcheon had a .030" plastic dial cover partially held in place with duct tape. The original dial covers were .090" thick plastic (cellulose acetate) and didn't need duct tape to hold them in place. The logging scale pointer was bent incorrectly to allow for a non-original mounting of the thin plastic dial cover. This had the pointer mounted with a screw and nut through the plastic of the dial cover. The original mounting had the plastic dial cover "notched" to provide clearance for the pointer which was then bent to have the pointer against the backside of the dial cover plastic. The pointer should be mounted to the escutcheon using a flat-head rivet.

A non-original hole was drilled into the escutcheon and into the lower part of the tuning shaft grounding fork to allow a screw and nut to mount the fork into a non-moving position for some reason. This was removed allowing the fork to be able to pivot on its mounting rivet. The drilled hole was filled with ABF and painted.

Weird Screw Sizes and Odd Rivets - Original escutcheon mounting screws are 4-40 oval head phillips machine screws 5/16" long. These had to be purchased. These screw heads had to be painted flat black before installation (originals were black oxide finish.) Original data plate mounting screws are 4-36 fillister head slotted machine screws 3/16"" long. I had six originals and needed two more. I made two fillister head 4-36 screws out of round head 4-36 screws. I turned the screw in a drill press and used a fine file to shape the round head into a fillister head.

The nomenclature plates on the dial escutcheon had been incorrectly mounted with screws and nuts. I had to order the correct 3/32" diameter by 1/8" length semi-tubular rivets to mount these plates as they originally were mounted. These rivets were vintage and had the correct "raven finish" (black.) I used a .125" sheet metal punch as a rivet set tool and a solid copper plate for the backing tool.

As mentioned, the logging scale pointer had been mounted with a screw and nut. The pointer itself needed to be reshaped correctly. It was then mounted as original with a flat-head rivet that was 3/32" diameter and 3/16" long. This was a solid aluminum rivet so a center punch was used as a setting tool and a copper plate used for the backing tool. 

All of the round nomenclature plates had been mounted with screws into tapped holes on the repro panel. On the original panel these had to be correctly mounted with #2 x 3/16" drive screws. The drive screws were ordered so these plates could be mounted as original. The drive screws were round head so before they could be installed the head had to be reshaped into a pan head. This was accomplished with a file. After the drive screws were installed their heads were painted black, as original. (I ordered round heads so these drive screws could be used on other restorations. You can always make a round head into a pan head but you can't go from a pan head to a round head.)

The bezel surrounding the band select dial was mounted with incorrect 4-40 screws on the repro panel. The screw holes in the original panel are tapped for 3-48 screws. The original screws were 3-48 flat head slotted machine screws. I found the local Ace Hardware had 3-48 FH screws but in phillips. Since the main tuning dial escutcheon is original-correctly mounted with oval head phillips screws, I decided to go ahead and use the 3-48 FH phillips. The heads of these screws had to be painted black before installation.

The original size 12-24 thumb screws can't be found as new items. I really couldn't find anything except original RAO-7 thumb screws that were the correct dimensions and would function as "captive" screws. Since the panel-mounted replica captive thumb screw standoffs were built exactly like the originals, if original RAO thumb screws are found, they can be fit into the standoffs. This isn't as easy as it sounds. Each RAO thumb screw has a boss just behind where the 12-24 threads stop. This is to retain a small "C" clip. The thumb screw is inserted into the captive screw standoff from the front. From behind the standoff, the "C" clip is installed onto the thumb screw threads and then pushed over the boss. The boss keeps the clip from sliding off of the threads and the clip is large enough to keep the thumb screw from being pulled out the front on the standoff.

The only solution is to have nine RAO thumb screw replicas made. Probably an expensive solution to the missing parts. For the time being, I can use the five RAO thumb screws that had been somewhat modified by a former owner and ruined for authentic restoration use. Although they will look original and function to hold the panel to the cabinet, they won't actually be "captive" thumb screws. 

Photo right shows the RAO-7 SN: J444 chassis and the appearance of a chassis that has had the MFP treatment. Note that obviously the tuning condenser wasn't coated. Most of this MFP is original. What I've done is to coat the cover on the antenna input-1st RF amplifier box which was not coated since it was an original-part replacement cover. Also, the AC power input receptacle was a NOS replacement and had to be MFP coated.

photo right: Close-up of the rounded cut-out for the tuning shaft and the notch that provides clearance for the logging dial pointer. Note the the pointer is bent to allow it to clear the dial cover yet lay flat against the back of the plastic. In both photos the protective plastic sheeting hasn't been removed from the Lexan.

photo above: left side handle and replica captive screw standoffs

The RF input box cover wasn't MFP'd. I thought it was replaced sometime in the past and it should have been coated when the receiver chassis was sprayed, so I went ahead and sprayed the cover with MFP. Now, I've seen some RAO-7 receiver chassis (both in person and in photos) where this cover is NOT MFP-coated but the rest of the chassis is coated. However, that would require temporarily covering the RF input box opening while the MFP is sprayed onto the chassis. Then installing the non-MFP'd cover. It doesn't seem likely,...but anything is possible.

photo right: MFP coated data plate

Crystal Filter Problem - When I attempted to find the crystal "peak" frequency for the filter I couldn't hear the usual abrupt signal increase that happens as the generator frequency is swept across 455kc. I suspected that the crystal was inactive for some reason. The left side panel has to be removed from the crystal filter assembly to access the components that are all located inside the metal box. The screws are all self-tapping hex head sheet metal screws. There is one stud mounting the panel to the chassis so that nut has to be removed from underneath the chassis. When I had access to the inside of the crystal filter I checked all of the connections to see if any were broken or were disconnected but everything looked good. I unsoldered the crystal which is the type that is entirely contained within a bakelite housing with terminals on the sides. I tested the crystal by "ringing" it. That is, connecting a RF signal generator, the crystal and an oscilloscope in series with the return being the 'scope ground to signal generator ground. At the crystal's active frequency, there should be a pronounced sharp increase in the signal shown on the 'scope. There was no "peak" at all. This indicated that the crystal was inactive for some reason.  >>>

RF Alignment - Tracking was very close on bands E, D, C and A. However, Range B was aligned to the image frequency and this required quite a bit of readjusting to get everything back to the correct tuned frequency. Alignment to an image is most likely to happen on Range A but the high end of Range B is over 14mc and with a sufficiently strong signal input from the RF generator it would be possible to hear the image and mistake it of the correct alignment frequency. When aligned to an image, the tuning range will not track properly (error at 10mc was about 500kc) and the gain across the range will not be equal. It's important, especially on Range A, to keep the signal generator level as low as possible. Also, check the location of the image,...it should be 910kc below the actual signal frequency input. To test the image location will require a fairly high signal input level to actually hear the image. If the image is at 910kc below, then that particular LO is adjusted correctly. It's also possible to adjust the Mixer to an image. As before, keep the signal generator level as low as possible to avoid mistaking the image for the correct frequency. A proper load should be used for the RF sections. Use a .1uf capacitor for Range E and a 100 ohm resistor for the other ranges.

Sensitivity was good, especially in the CW mode. The AVC is very strong and cuts back the sensitivity quite a bit in the AVC mode which is how most AM listening is done. Selectivity is about 8kc bandwidth without the crystal filter and can be narrowed as needed with the filter. Audio is pretty good considering the restrictions that the USN wanted in the receiver. With a large, matched speaker, AM and SW BC stations sound very good. 

Since the RAO-7 was specifically for use with a panoramic adaptor, most of the RAO-7s were operated in MVC or BFO. If the receiver were in AVC then the center peak signal on the panoramic adaptor would be attenuated due to the AVC action cutting back the sensitivity of the receiver. The panoramic adaptor circuitry does increase the gain at the edges of the bandwidth sweep and that gives the impression that signals 20kc or 30kc off center are much stronger than the "tuned signal." Using MVC on the RAO-7 results the panoramic display showing relative signal strengths more accurately. Or, if you know why the tuned signal appears reduced in amplitude, then just use AVC - especially if you're operating in Net-type communications - and enjoy the panoramic display of the adjacent frequency activity - aka: QRM!

If the RAO-7 was used for communications, then that most likely would have been in the CW mode. In its WWII use, most RAO-7 receivers probably were very seldom used with the AVC turned on. However, today most RAO-7 receivers will probably be used in the AM mode with AVC on. Expect the sensitivity to be somewhat restricted by the "strong" AVC-action. Typically, for AM, RF gain will be at 9+ and the AF gain will be set to about 5 for loudspeaker listening. If the AM sounds a bit rough, just reduce the RF Gain a little. Phones are connected directly to the output transformer so the AF gain would have to be reduced significantly for headset listening. You can also connect the 600Z ohm loudspeaker to the Phones jack if you don't want to use the rear chassis terminals.

The RAO-7 doesn't have remote standby capabilities due to its intended use during WWII. In fact, none of the RAO receivers have remote standby. It's easy to just use the front panel B+ ON-OFF switch for the standby function. Receiver antenna input isolation must be considered and this can be accomplished with either an external DowKey relay or by using the Sending relay internal to most commonly used vintage military transmitters. Separate transmit and receive antennas could also be used.

photo above: Dual RAOs  -  RAO-7 sn: H720 above RAO-7 sn: J444 below
 

I had my two as a "pair" and "paired with" an AC operated ART-13 transmitter. Certainly not protocol during WWII but I did see in a 1948 Radio News magazine where a ham station in Hawaii was so equipped. Nice receivers that, although they weigh a lot, also provide a lot of performance not to mention their incredible bench presence.

Nowadays, that is 2021, the RAO-7 H720 has been paired with the Navy RCX Panoramic Adaptor SN:46 and operated with the Navy-Collins T-47/ART-13 transmitter. RAO-7 J444 is set up in the shop and can operate with the ART-13A located there.

In looking at the photo to the left, how did I ever get H720 on top of J444? In moving the RAO-7, it's very easy to pull the receiver out of the cabinet. This reduces the weight by about 20 pounds,...down to about 55 pounds for the chassis. I put the cabinet on top of J444 first and then installed the receiver into the cabinet. Most "heavy lifting" can be avoided by some form of disassembly and, in this case, disassembly was also necessary when I moved H720 off of J444.

Shown to the right is that photo from Radio News - May 1948 showing KH6DA Lt.Cmdr. H.E. Warren ret. (in the middle) at his station in Honolulu in 1948. The ham station consisted of the RAO-7 receiver and to the right is the ART-13 transmitter. Actually, the article in Radio News was about KH6DA using his ham station for a net operation for a meeting of a "camera club" that he belonged to.

I've operated two RAO receivers that have been rebuilt, my just recently completed RAO-3 and a RAO-7 that had been rebuilt by a fellow ham back in the 1990s. As a comparison, I've also operated my "all original" RAO-7 sn: H720 quite a lot. While H720 is usable and functions fine, the two rebuilt RAO receivers do have better audio and much better RF tracking. Could H720 be improved with just an alignment? Probably. One has to consider the rarity of the "all original" aspect of H720 and accept that its performance "is what it is." But an alignment is non-invasive and generally improves performance (especially if the last alignment was over 75 years ago.)

Why aren't the RAO receivers more popular with hams operating vintage military radio stations? I only know of two other hams that use or have used RAOs as their station receiver. N6MKC used a RAO-6 paired with a TKC-7 transmitter and NU6F uses a rebuilt RAO-7 (the one that was rebuilt in the 1990s) once in a while as his station receiver. I've used both my RAO-3 sn:1064 and RAO-7 sn:H720 as station receivers. All vintage military radio operations are on 75M and that doesn't present any challenges for the RAO. Adjacent frequency activity is about the only issue on 75M and the RAO crystal filter can handle that with ease. So, performance probably isn't the reason for lack of popularity. History might be a factor. Almost all RAO receivers were used for intercept, surveillance or relay communications. A few were on ships and submarines but the romance of WWII shipboard operation probably can't be attributed to most RAO receivers. Weight versus performance is probably the primary complaint. At 75 pounds, the RAO is a challenge to move around. Communications grade audio with the lack of robust bass response usually isn't a concern for vintage military radio collectors but it might be factor when combined with the hefty weight of the RAO.

Maybe I have a "soft spot" for the RAO but I think it's a great receiver and an accurately rebuilt RAO is a pleasure to operate and listen to.

US Navy RAO-3 by Wells Gardner & Company - 1943
 

U.S. Navy "Catalogue of Electronic Equipment" Volume 2 (NAVSHIPS 900,116) - Detailed descriptions of all RAO models.

Radio News magazines from 1946 up to 1950 have lots of advertising from military surplus dealers showing what was available and the prices then

Fair Radio Sales catalogs from the 1960s, a little bit later but also show what military surplus items were available and the price then

QST magazine ads 1930s to 1950s for NC-100 history and coil catacomb information some info on WWII RAO in photos in wartime issues

Online - www.navy-radio.com  Nick England's fantastic compilation of Navy radio history and photos. Many photographs of different RAO receiver versions. Lots of information.

Bama Edebris - Boatanchor Manual Archive online manuals allows reading and checking data without having to buy a copy of a manual. Of course, you can download and print if you need a hardcopy.