Radio Boulevard
Western Historic Radio Museum
 

Collins Radio Company

51J Series of Communications Receivers

PART 1: History, Design and Circuit Details for the 51J-1, 51J-2, R-388/URR,
51J-3, 51J-4, 51J-5 prototype, 51S-1, Lab 51J-4,
Beckman Model 7700 (51J-4) and several other 51J variations, 51J Cabinets 
How Collins Radio and the 51J were Responsible for the RACAL RA-17
 
PART 2: 51J "Oddities," 70E-15 "M" PTO, BFO Inversions, Break-in Set ups,
Details on Common Problems, Repairing the 70E-15 "M" PTO, Rebuilding the 51J-2,
 Rebuilding the "Basket Case" R-388, 1952 R-388 in CY-1260/G Military Receiver Case

PART 3: Refurbishing the Light Gray Panel 51J-4, Restoring the 51J-1,
 Rebuilding the R-388 w/ 354A-1 Mechanical Filter Kit Installation
 

by: Henry Rogers WA7YBS-WHRM/Radio Boulevard

Artwork from the 51J-4 Advertising Brochure

51J Series of Communications Receivers - Part 1

Introduced through advertising in 1949, the Collins 51J receiver was intended for commercial-professional users that required an extremely accurate frequency readout and a very stable, drift-free receiver that was especially suited for use in data reception, such as RTTY, but could also provide excellent communications reproduction of AM or CW signals. The 51J was also the perfect receiver for a wealthy ham or SWL enthusiast - if he could afford its nearly $900 price tag. Collins Radio Company had entered into the ham radio receiver market with their ham bands only receiver, the 75A, two years earlier. But, according to Collins Radio (the CCA actually,) the development of a precision general coverage receiver had started (at Collins) towards the end of WWII, in 1944. The project lead was Roy Olsen from 1944 up to 1946 and then Lou Cuillard took over until completion (but no date is supplied as to when "completion" happened.) The new receiver was designated the 51J and the intended market was post-WWII commercial-professional users that could afford an expensive but advanced-design receiver. It's possible that the general coverage 51J was available in very limited numbers to a few commercial or military users in 1947. The number of 51J-1 receivers produced from 1947 through early-1948 was an extremely small quantity and was probably only for testing and possibly for samples that were provided to the Signal Corps for their testing and evaluation. Most of the 51J-1 production, small as it was, happened from mid-1948 through all of 1949 and into very early 1950. The 51J would provide commercial and possibly military users with general coverage reception from .5mc up to 30.5mc using thirty, 1.0mc wide tuning ranges and, with its crystal oscillator and permeability-tuned oscillator conversions providing better than 1kc dial accuracy and drift-free stability, the receiver itself featured performance that was literally years ahead of the competition. 

Signal Corps and Early RTTY Receivers - During the post-WWII era, the U.S. Army Signal Corps was trying to use some of the WWII vintage receivers for RTTY (Radio Teletype) and were finding the drift to be excessive for RTTY and the dial accuracy along with the rapid tuning rates to be completely inadequate. These receivers were usually the BC-342 in a highly modified form and the BC-779 Super Pro, also highly modified from its WWII form. Crystal control of the LO and the BFO were necessary on these receivers in order to have sufficient stability for RTTY capability. The story goes that Collins sent an early 51J receiver to the Signal Corps to try out. This certainly was before the 51J was commercially released. The Signal Corps' response was positive since the receiver was extremely accurate in its frequency readout, its tuning rate allowed easy tuning of very narrow signals and there was essentially "no drift" in the Collins receiver in stock form. The Signal Corps "had to have 'em" and soon Collins was providing the Signal Corps with the R-381/URR and the R-381A/URR, those receivers being the 51J-1 and 51J-2 respectively. However, there were some non-military aspects to these initial 51J designs that the Signal Corps wanted changed. The initial 51J had a fixed 300Z antenna impedance that really didn't fit many of the military antenna types. It also had the hams bands highlighted in green on the MC tuning dial implying a civilian market. It seemed that Collins had a receiver designed for commercial users or maybe a few wealthy hams but had also considered that the precision tuning and stability of the design would easily meet the military requirements necessary for reliable RTTY. So, although the initial 51J receiver as the R-381 and R-381A provided the Signal Corps with a thoroughly modern design that had the most of the RTTY specifications the had military sought, some minor "tweaking" was necessary for the receiver to become the final evolution that the Signal Corps and other military users needed. That version became the R-388/URR. Many thousands of R-388/URR receivers were built by Collins from 1950 up to about 1962. The R-388 was found in many frequency diversity RTTY set-ups used by the military, many times installed in portable RTTY communications huts.

Hallicrafters R-274

Hallicrafters R-274 receiver was first produced in 1949. Of course, the design was almost all Signal Corps work with some Hallicrafters input. The Signal Corps wasn't impressed with the final product (even though it was basically their design) and only one contract was issued for the R-274. A second contract was issued for its updated successor, the R-274D, about 1952. Hallicrafters was probably planning on greater success for the R-274 and had certainly tooled for that type of production. It ended with Hallicrafters offering the R-274D as a ham receiver, the SX-73, but not many hams were willing to pay the $900+ asking price (that's over $11,000 in 2023 dollars.) The R-274 was a fine general purpose receiver,...for 1949,...just not for RTTY.

Hammarlund SP-600JX

The SP-600JX also wasn't very good as a RTTY receiver but the the Army and the Navy (and ultimately the USAF) all ordered SP-600JX receivers by the thousands as a general purpose receiver. The SP-600JX version with the six-channel Crystal Oscillator was intended for increased stability for RTTY operation. The receiver was very difficult to keep on frequency unless the Crystal Oscillator was utilized. That required knowing what frequency was needed and then calculating the crystal frequency and then procuring that crystal. Then that could be repeated five more times to have the full six-channel capability for Crystal Oscillator control. The SP-600 was a fine general purpose receiver and ultimately became popular as a surveillance receiver because of its wide frequency coverage per band. The Hallicrafters R-274 also featured this type of six-channel Crystal Oscillator option.

In both the Hallicrafters R-274 and Hammarlund SP-600 receivers, each tuning range selected covered wide segments of the spectrum and that left the dial resolution vague and the tuning rate fairly rapid. Knowing your tuned frequency left a lot to the imagination and an accurate measurement of the tuned frequency required using a heterodyne frequency meter. While the Crystal Oscillator option did reduce the drift in the LO frequency in both of the receivers, either one still had to be manually tuned to the selected tuned frequency so that the RF and Mixer stages would be properly tuned to the desired frequency as well. It was all such a hassle that when Collins submitted the 51J receiver (and later the R-390 receiver) the difference in technological design superiority was so obvious that the Signal Corps really couldn't make any other choice for successful RTTY, along with all other types of communications and surveillance, than to use the Collins 51J receivers (and later, the R-390 receivers.)

Several 51J-4 receivers were produced for commercial laboratories like Beckman. Beckman produced special Frequency Measuring Systems that had a modified 51J-4 working with other special equipment that Beckman produced. The 51J-4 receivers produced for Beckman have special panels with "Beckman" or "Beckman/Berkeley" in the upper left panel area. Also, most Beckman 51J-4s are identified as "Receiver Model 7700" and have a light-gray finish on the front panel. Additionally, some 51J-4 receivers were supplied to other commercial laboratories and test facilities with light-gray front panels. Even Collins' own laboratories and test set-ups used 51J-4 receivers with the light-gray semi-gloss front panels. 

The 51J-4 was also found in overseas embassies, at universities like Stanford, even some wealthy SWLs (supposedly comedian/actor Jackie Gleason was an avid SWL who owned a 51J-4) and maybe even some ham enthusiasts bought 51J-4 receivers, though the $1200 purchase price limited that market. The 51J-4 was in production until 1964 with a production of at least 7500 receivers (that's about how high the 51J-4 serial numbers go.) All total, with the early 51Js, the R-388s and the 51J-4 included together, the production probably was over 20,000 receivers.

KPH Point Reyes operator Fred Baxter with the 51J-4

51J Circuit and Design Description

Basic Circuit Design of All 51J Receivers

The Front End - All 51J receivers utilize the same basic front end, that is, a permeability tuned, double conversion circuit that starts with the incoming RF signal being mixed with a selectable multiple frequency crystal oscillator to produce an output that can be tuned in a 1mc wide range of either 1.5-2.5mc for Even bands or 2.5-3.5mc for Odd bands. These Odd or Even bands comprise the dual Variable IF of the 51J circuit. The output of the First Mixer combines with the proper selected Odd or Even Variable IF to become an input to the Second Mixer. The other input to the Second Mixer comes from the 3.0mc to 2.0mc VFO (called a Permeability Tuned Oscillator or PTO) and the Second Mixer output produces a fixed 500kc IF signal.

The band select numbers start with .5 to 1.5mc being Band One (odd,) 1.5 to 2.5mc being Band Two (even,) etc., on up to 29.5 to 30.5mc being Band 30 (even.) Band 1 actually is triple conversion, using the "Band One Mixer" tube but that mixer is only used for Band 1 and was only necessary to provide coverage of the AM BC band. Bands 2 and 3 are single conversion since they are essentially the frequency coverage of the dual Variable IF system which is used as the RF amplifier for Bands 2 and 3 with the output going directly to the Second Mixer (mixing with PTO output to produce the 500kc fixed IF.) Due to the single conversion on Bands 2 and 3, the tuning is the reverse of the double and triple conversion bands. Bands 2 and 3 have their MC scales in red (with highest frequency to the left of the scale) and the red scale on KC dial tracks the opposite of the black KC scale. All of the remaining bands are double conversion and select alternating Even or Odd tunable IF output circuits as required. The selected Crystal Oscillator and RF amplifier band will provide the correct 1.0mc-wide tuning range to the proper selected Variable IF and that output then can mix at the Second Mixer with the 3.0mc to 2.0mc tuning range of the PTO to provide a fixed-500kc output to the IF. By using two selectable and variable IFs, the required number of crystals for the Crystal Oscillator to cover 28 bands (Bands 2 and 3 are single conversion and don't use the Crystal Oscillator) can be reduced to just ten crystals if the fundamentals and harmonics are used. This is the basic 30 band front-end that all 51J receivers employed.

Since the dual Variable IF presents the Even bands lower in frequency than the PTO output and Odd bands higher in frequency than the PTO, an "inversion" of the input to the fixed 500kc IF happens. This is a transparent function when the BFO isn't employed meaning it isn't noticed in the AM mode. Even CW, with its lack of sidebands, wasn't affected. However, as single sideband transmissions developed and started becoming the dominate voice mode, the IF inversion between Even and Odd bands was noticed because it did affect whether the BFO was set higher or lower in frequency related to the 500kc IF as the method to select the proper sideband being transmitted and that BFO position changed depending on whether the band in use was Odd or Even. This "inversion effect" really couldn't be changed since its cause was due to the basic design of the receiver's front end and all 51J Series receivers will behave in this manner.

The 500kc Fixed IF - The 500kc fixed IF remained basically the same for all versions except the 51J-4. Three stages of IF amplification were provided with a Crystal Filter that had four steps of Selectivity and Phasing controls. The Crystal Filter circuit was identical in all 51J receivers. The 51J-4 receiver added three mechanical filters to the IF with the installation of the 354A-1 Mechanical Filter Kit at the production level. The mechanical filter insertion-loss required an additional IF stage so there were four IF amplifiers in the 51J-4. The IF transformer primary and secondary windings were fairly loose-coupled and that resulted in a narrow IF bandwidth of about 4kc at -6db in the 51J-1 and 51J-2 receivers. The R-388 added 2pf coupling capacitors across the primary and secondary to slightly widen the bandwidth to about 6kc at -6db. The 51J-4 increased the coupling capacitors to 3pf because the mechanical filter selected determined the bandwidth.

The Det, AVC, NL and Audio Sections - In all receiver versions, the detector is a diode-connected dual triode 12AX7 with the second diode-connected triode used as the AVC rectifier. The AVC amplifier was a parallel-connected dual triode 12AX7 in the 51J-1 and J-2 but, when an IF output was required for the R-388, this tube was changed to a 12AU7 and then one triode was the AVC Amplifier and the other triode was the IF output buffer. The Noise Limiter circuit was the same in all versions and so was the BFO (other than a couple of NPO caps added in the R-388 BFO to eliminate any f drift.) The audio output tube was always a 6AQ5 but its location was in the back corner of the chassis on the 51J-1 and early 51J-2. Its location was moved next to the 5V4 rectifier tube during 51J-2 production and remained in that location for all other subsequent 51J versions. The power supply for the 51J-1 and 51J-2 used a 5V4G rectifier and didn't have any voltage regulator. The 0A2 regulator tube was added beginning with the R-388 and 51J-3 and continuing on in the 51J-4. Also, changed with the R-388 was the type of PTO used. 51J-1 and 51J-2 receivers used the 70E-7A PTO with single 6SJ7 oscillator tube while the R-388, 51J-3 and J-4 used the 70E-15 PTO with the 6BA6 VFO tube and 6BA6 VFO Buffer stage. The 100kc Calibration Oscillator always a crystal-controlled oscillator that was connected to the grid of the RF Amplifier tube in all versions. All versions had a 4.0Z ohm audio output and a 500Z was provided in the 51J-1 and 51J-2 but that was changed to a 600Z ohm audio output in all subsequent versions. The Stand By function in the 51J-1 and 51J-2 parallels the front panel Stand By switch. The R-388, 51J-3 and J-4 use a Break-in relay that isolates the antenna, grounds the receiver input and removes the IF plate voltage.

Details for Each 51J Model

Collins had been developing the 51J general coverage receiver since 1944. It seems unlikely that the 51J-1 receiver was available in 1946, concurrent with the introduction of their ham receiver, the 75A (to become the 1947 75A-1 when the noise limiter was added.) It's probable that a few prototype 51J-1 receivers were being developed into marketable receivers by 1947. Sometime in the late-1940s, Collins sent a few 51J receivers to the Signal Corps for testing and evaluation. The first advertisements appear in 1949. Since the 51J-1 was the introductory version, several changes happened along its very limited production history that more than likely ran from mid-1948 through all of 1949 and up to early 1950 when its successor was introduced, the 51J-2 which was available through all of 1950.    >>>

>>>   Since the initial design was from the end of WWII, the typical receiver antenna inputs at that time were fairly high impedance. All 51J-1 versions had an antenna input impedance that was fixed at 300 ohms. This required the user to "match" the antenna impedance to the receiver input impedance for best performance. This could be using a resonant folded-dipole antenna or using an antenna tuner for other types of antennae. The 51J-1 didn't provide an antenna trimmer circuit to compensate for varying input impedances since the designers expected the end-user to provide the "matching" necessary for best performance. The 300 ohm antenna impedance was due to the ANT/RF coils being actual RF transformers and having primary windings. This provided excellent front-end selectivity but also complicated the types of antennae that some end-users wanted to employ. Vertical whips and dipole antennae were much lower impedance types while random end-fed wires could vary impedance depending on frequency in use but usually presented a very high impedance. For top performance, all of the antennae that the military or commercial users would have employed would have required an extra matching device when using the 51J-1 and 51J-2 receivers. The Antenna input connections were to a three screw-terminal strip marked A1, A2 and G mounted on the rear chassis.

The PTO used was the 70E-7A that had a single 6SJ7 metal octal tube as the oscillator. The 6SJ7 tube was covered by a screw-mounted cylindrical metal shield. The "easy to access" adjustment for end-point error was located on top of the PTO under a removable triangular metal cover plate. The glass envelope in front of the 6SJ7 contained desiccant crystals that changed color as they absorbed moisture going from blue to pink. When the desiccant was pink it was supposed to be replaced with new desiccant to continue the absorbing moisture action, though the replacement seldom happened. The fixed IF was tuned to 500kc and had three amplifier stages. A Crystal Filter provided four switched steps of selectivity and heterodyne interference reduction using the PHASING control. A standard envelope detector, AVC amplifier, AVC rectifier and a Noise Limiter were also included. 16 tubes were used in the 51J-1.

The 51J-1 front panel featured a black-painted metal dial bezel with "51J RECEIVER" silk-screened in white, the Collins' chrome and orange "winged emblem" and no grab handles. The S-meter was just that,...a S-meter that was very similar (if not identical) to the S-meter used on the 75A-1 receivers. The S-meter was illuminated and had a light-yellow scale, though the material was highly photosensitive and rapidly darkened to an amber color. Later, as more 51J-1 receivers were sold to the military (as the R-381,) it's possible that the S-meter was replaced with a combination Carrier Level and Audio Output meter with a white, dual scale that was illuminated, as was installed on the 51J-2. It's also possible that these later style meters and the associated Meter switch were merely upgrades incorporated into the R-381 receivers.

The BAND CHANGE knob had a crank handle that was mounted at the outer edge of the knob. It's possible that the crank style changed during production (see "pull-out" crank on the 51J-2 further down.) The CALIBRATE nomenclature was originally going to be 100KC CRYSTAL and that was on the initial prototype 51J-1 receivers that were used for artwork that was used in advertising (see 1949 ad in History section above.) When the 51J-1 went into production the 100KC CRYSTAL nomenclature was changed to CALIBRATE (I've never seen any 51J-1 receiver that had the 100KC Crystal nomenclature.) The CAL. was a fine adjustment trimmer for the 100kc crystal oscillator (the wide-range adjustment was on the chassis.) ZERO ADJ. was a mechanical adjustment of the KC dial fiducial index for precise frequency readouts or measurements.

The 51J-1 audio response was "communications grade" with a low end of 200hz (at -6db) and a high end of 2500Hz (at -6db.) A .01uf capacitor was installed on the primary of the audio output transformer for impedance matching and transient suppression and this tended to "roll off" the high frequency response and that left the audio sounding somewhat "muffled" when listening to voice or music transmissions. However, most of the restricted audio response is due to the IF section and its very narrow bandwidth of 4 to 5kc at -6db down. Except for changing the value of the .01uf capacitor to 6800pf in the R-388, which resulted in slightly different specifications (200hz at -3db to 2500hz at -7db,) no significant changes were made to the audio output stage for the entire 51J Series (but, also starting with the R-388/51J-3, small value capacitors of 2pf were installed across the IF transformers, primary to secondary, to increase coupling and that also broadened the bandwidth of the IF section.) The 51J-1 and J-2 receivers can be tuned to either the upper or lower sideband when in the AM mode to increase the audio highs. The audio output transformer provided a 4 ohm impedance audio output and a 500 ohm impedance audio output. With the R-388/51J-3 the 500Z was changed to 600Z. The loudspeaker described in the manual has the dimensions and specifications of the Collins 270G-1 though it isn't identified with its model number.

The power transformer used for the 51J-1 was a frame-type, vertical mount transformer. The frame-type power transformer was painted gray. This frame-type was changed to a potted power transformer by the 51J-2 production. The audio output transformer also was a frame-type vertical mount that was changed to a potted transformer by the 51J-2 production. Remote Standby was accessed from a terminal strip on the rear chassis and parallels the front panel Standby switch position. The ham bands were high-lighted in green on the megacycle drum dial but, at $875 list price, not many hams could afford a 51J-1 as their station receiver (if you wonder why few hams bought the 51J-1, $875 in 1949 was equivalent to $11,150 today in 2023.) Incidentally, the $875 price did include a table top cabinet for the receiver, if so ordered. Otherwise, a rack mount configuration was supplied.

The 51J-1's chassis layout is slightly different when compared to the later receivers. The most obvious difference is the four tubes in-line at the left-rear corner area of the chassis (viewed from the front of the receiver.) On the 51J-1 and early 51J-2, the IF transformers were in-line, front-to-back, but were more forward on the chassis. This allowed space for the four in-line tubes that comprised the Detector, AVC, NL and Audio Output functions. The tubes used were three 12AX7 tubes and then the left-most tube was the 6AQ5 audio output tube. Also, this layout resulted in the 1st IF amplifier tube (6BA6) being located directly under the MC drum dial which made removal of that tube difficult. This layout was changed during the 51J-2 production (beginning with 51J-2 sn: 501.) Another difference is the location of the fuse holder next to the 5V4G rectifier tube. This fuse location was changed during the 51J-2 to the more conventional rear chassis apron location. Also, the filter capacitor is a fiber-board mounted, multi-section electrolytic capacitor (the plug-in filter capacitor was introduced with the 51J-2.) There may have been a black cardboard tubular cover over the filter capacitor. Yet another difference was the crystal oscillator tube (V-5, later V105) that originally was a 6BA6 but was changed to a 6AK5 during the 51J-2 production. A close-up photo to the lower left shows the "four tubes in-line" left rear corner of the receiver chassis along with the fuse holder mounted next to the 5V4G tube.

51J-1 Chassis Top - The 70E-7A PTO is in the center of the chassis just in back of the MC drum dial (note the green highlighted 80M band.) Also, note how close the 1st IF transformer is to the front of the chassis when compared to later receivers. This made removal of the 1st IF amplifier tube quite difficult since it's almost directly under the drum dial.

51J-1 - Rear-left top of the chassis showing "four tubes inline" and the top of the chassis fuse holder next to the 5V4G tube. Also, note that the multi-section filter capacitor is a fiberboard chassis-mounted type, not a plug-in.
 

51J-1 - Under chassis showing the "half-shield" and note that the two chokes are frame-type units.

The photo to the right shows the serial number area under UV. The "6" is fairly obvious but the other digit is all but gone. I put the image into "Photoshop" and there I could enlarge it and use different filters to bring out the second digit. Although very faint, with the enlargement capabilities of "Photoshop," the "top half" of a "5" can be seen. The idea of photographing the serial number under UV light and then uploading into "Photoshop" was a major help since I could manipulate the image for best results. It's also very likely that different wavelengths of UV might show ink residue better. You do have to be careful when using some wavelengths UV, don't look at the UV light directly and wear UV protecting glasses. Also, don't look at reflected UV light for long periods of time. Just take a photograph and then examine the results in "Photoshop."

UV light + Photoshop Reveals Serial Number

The 51J-2 was the slightly evolved version of the 51J-1 that was now ready for "full on" production. The 51J-1 limited production had resulted in "fine tuning" the manufacturing of the 51J receiver resulting in a reliable, high performance commercial-military receiver that could also be used by wealthy hams or SWLs. With the 51J-2, production levels were increased dramatically and this was because by early 1950, when the 51J-2 was introduced, the military had become interested in the 51J and were ordering the R-381 and R-381A versions of the 51J-1 and 51J-2. Despite a production period that was just under one year, the 51J-2 seems to have been produced in much larger quantities than its predecessor. Evidence indicates that the 51J-2 production exceeded 2300 receivers in a time period of less than one year.

Like the J-1, the 51J-2 versions had an antenna input impedance that was fixed at 300 ohms. The PTO used was the 70E-7A that had a single 6SJ7 metal octal tube as the oscillator. The fixed IF was tuned to 500kc and had three amplifiers. A Crystal Filter was provided along with a standard envelope detector, AVC amplifier/rectifier, Noise Limiter and Audio Output section. 16 tubes were used in the 51J-2.

The 51J-2 receiver had the same black-painted metal dial bezel with "51J RECEIVER" silk-screened in white, the Collins' chrome and orange "winged emblem" and no grab handles. The S-meter was changed to an illuminated, square-front bezel Carrier Level and Audio Level meter. Audio Output level measuring function was actuated by a toggle switch located next to the meter. The new dual scale Carrier Level/Audio Output meter had OUTPUT above the upper scale and INPUT RF below the lower scale. The scale itself was white. This METER switch and dual scale meter were the most apparent differences between the 51J-1 and J-2.

Audio response remained the same as the 51J-1 and was definitely not high fidelity, usually producing "muffled" audio highs when receiving AM voice signals. Like the 51J-1, the primary cause was the very selective IF transformers that resulted in the 4kc at -6db IF passband. Audio output impedances were 4Z and 500Z. The loudspeaker described in the J-2 manual fits the 270G-1 parameters for size and impedance.
 

51J-2 - Chassis after sn:500 - Note the potted power transformer and the relocated 6AQ5 tube and the rear chassis apron fuse holder. Also note that the multi-section filter cap is the plug-in style.

Like the 51J-1, the REMOTE terminal strip on the back chassis apron provided a stand by function that was in parallel with the STAND BY on the front panel. It might be possible that the green high-lighting of the ham bands on the megacycle drum dial was eliminated in the later 51J-2 receivers since it was unlikely that sales to hams accounted for many purchases. But, later rebuilds probably account for receivers found with R-388 MC dial drums installed. (51J-2 sn:2273 still had the green highlighting on its MC dial drum.) The power transformer and the two filter chokes were changed to potted units to increase reliability. The wiring harness changed to vinyl insulated wires during the 51J-2 production. The AC power cord on the 51J-1 and 51J-2 receivers was originally a two-conductor black rubber "zip cord" with a period molded AC plug. Nearly always, the original AC cords have deteriorated long ago and were replaced with the heavy-duty round power cables in both two-conductor or three conductor varieties commonly found on these receivers today. 

The photo to the left shows the chassis of the later 51J-2 (with serial numbers higher than 500) which has a slightly different layout than the first 500 receivers that were like the 51J-1 in that regard. Beginning with 51J-2 serial number 501, the 6AQ5 was relocated next to the 5V4G rectifier tube and the fuse holder moved to the rear chassis apron below the AC power cable location. This location for the 6AQ5 and the fuse holder continued for the remaining J-2 production and all subsequent 51J and R-388 receivers. V-105, the Crystal Oscillator tube had been a 6BA6 for all of the 51J-1 production and for some portion of the 51J-2 production. During the J-2 production run, V-105 was changed to a 6AK5 tube. These later-built receivers will have "6AK5" ink-stamped on the chassis rather than a silk-screened ID. The filter capacitor was changed from the fiber-board mounted multi-section used in the 51J-1 to an easy-to-replace, plug-in multi-section unit, sometime during the 51J-2 production.

51J-2 - The characteristic "sloping top cover" of the 51J Series. All 51J receivers will have this type of top cover. The bottom three retaining fasteners were screws on the J-1 and J-2 receivers. R-388 and later receivers had studs and wing nuts to fasten the back of the top cover. The serial number plate isn't original (it covers a non-original SO-239 mounting through the original silk-screened ID.)

The 51J-2 was built through all of 1950 and was replaced with the introduction of the military R-388 and the civilian 51J-3 towards the end of 1950. When the military ordered both 51J-1 and 51J-2 receivers they were designated as the R-381/URR for the 51J-1 and R-381A/URR for the 51J-2. Some 51J-1 and 51J-2 receivers will be found without the Collins' Winged Emblem mounted on the front panel but the receiver panel will be drilled for installation of a data plate, which, of course, is nearly always missing. The implication is that civilian 51J-2 receivers had the Collins WE while the R-381A versions had a military data plate installed instead. This also applies to the 51J-1 versus the R-381/URR military versions of that receiver.

There are indications in Collins manuals and other documentation that the 51J-1 and 51J-2 did carry serial numbers, however, I've never seen a serial number in the location provided on the rear chassis apron (or anywhere else on the receiver.) All examples I've inspected of both the 51J-1 and the 51J-2 receivers always are missing the serial number. Collins did use a lot of "ink-stamped" numbers in their production so it's possible that the serial number was ink-stamped using water-soluble ink and the stamping just hasn't survived (see updates below.) Even though the 51J-2 production only lasted for about one year, it's evident that in that short time, several times as many 51J-2 receivers were produced than the total quantity of 51J-1 receivers produced from mid-1948 up into early 1950.

51J-2 Under chassis "half-shield" - Note the potted power supply chokes used in the J-2

IMPORTANT UPDATE: May 23, 2023 - Retrieving Ink-Stamped Serial Numbers -  A 51J-2 "parts set" that was listed on eBay had a close-up photo of the rear chassis apron. Just visible was the "ink-stamped" serial number of "2273." This confirms that indeed the serial numbers on 51J-1 and 51J-2 receivers were ink-stamped, probably in water-soluble ink, in the area provided on the rear chassis apron. It appears that the ink just doesn't survive any type of cleaning or it might not even survive a humid environment. Also, the serial number "2273," if sequential as they should be, indicates a fairly large production quantity of 51J-2 receivers.

See "update" in the 51J-1 section above regarding using a UV flashlight or other UV light source to recover the original serial number by illuminating the ink residue. However, I didn't have anyway to put SN:2273 under a UV light, so I tried using Photoshop to enhance one of the eBay photos. Using enlargements and filters, I was able to bring out all of the numbers that comprised the serial number. UV would have been better, but this method also works if the ink isn't too deteriorated.

Interesting Note: V-105 6BA6 versus 6AK5 and Serial Numbers - The 51J-2 profiled above was modified in the past to install a SO-239 connector. The installation was right in the ID-SN silk screening on the rear chassis apron. I installed a cover-plate over the modification's holes that I made using the silk-screened aluminum area that I cut-out the derelict 51J-2 "parts set" chassis. Using the UV light, I was able to determine that the ID cover-plate I made had a serial number of "1653" just barely visible as ink residue when illuminated with a UV light. But, what about the original SN under the cover-plate? I didn't remember if the SN area was destroyed or not so I removed the plate to see. Fortunately, the actual SN area had not been drilled. Under UV light I could just make out the SN was "755" in the ink residue. The 51J-2 shown above and the parts set 51J-2 have the 6AK5 installed for V-105 and both chassis are ink-stamped "6AK5." However, from photos of the chassis of 51J-2 SN:2273, I could see that V-105 was a 6BA6 with that silk-screened ID. What is interesting is that both SN:1653 and SN:755 have the 6AK5 V-105 update while the later receiver, SN:2273, still has the 6BA6 V-105. The explanation might be the Collins production method of building ahead, placing "complete but not finalized" receivers into stock and subsequently removing those receivers from stock to fulfill orders at which time the receiver would be final tested, QA'd and serialized. The stocking process was apparently an unorganized quantity of built (but not yet serialized) receivers and that might explain how the later SN receiver still has the earlier V-105 tube. Also, there are reports that sometimes Collins in-house testing would require a receiver which would be pulled from stock, used for an indeterminate time period for testing and then subsequently returned to stock. Either of these scenarios could explain the "6BA6 V-105 showing up in a later SN receiver."

Enhanced Photo of SN:2273

NOTE on Information Source: 51J-2 SN:501 Changes - The information on the major chassis change from the earlier 51J-1 (with the "four tubes inline") to the 51J-2 layout happening with SN: 501 is implied in the Collins 354A-1 Mechanical Filter Kit manual. That manual is very specific that the 354A-1 can only be installed into 51J-2 receivers with serial numbers of 501 and higher. This was due to the IF transformers of the 51J-1 being closer to the front of the chassis and that changed the available space dimensions, meaning the 354A-1 literally wouldn't fit into 51J-1 or into 51J-2 receivers with serial numbers of 500 or lower. Note with the installation of a 354A-1 into a R-388 receiver shown further down this page that the 354A-1 kit is almost up against the MC dial drum, so the installation barely fits into the R-388/51J-4 receivers and simply could not fit into a 51J-1 or early J-2.

Towards the end of 1950, the R-388/URR was introduced, featuring an 18 tube circuit (adding a voltage regulator 0A2 and VFO buffer 6BA6,) a new version of the PTO (70E-15) and eliminating the fixed 300 Z ohm antenna input (by removing the primary winding on the antenna coils) and redesigning the antenna input to a more flexible design utilizing an Antenna Trim control. This revision was probably at the request of the Signal Corps, who wanted to use the new version of the receiver for their RTTY installations but found old 51J antenna requirements of a fairly Hi-Z, "fixed" (non-adjustable) 300 ohm antenna input impedance difficult to work with since most of the Army installations used either 75 Z ohm dipoles or 50 Z ohm (or less) vertical whip antennas. The new upgraded receiver was designated as the R-388/URR (when going to the Signal Corps) and it was built from late-1950 through about 1954 in relatively large quantities. There were also later contracts for 1955, 1956, 1957 and 1962 but these later contract quantities only total about 1000 receivers. It's commonly believed that from 1950 through 1962 somewhere around 12,000 R-388/URR receivers were produced. PTO serial numbers imply that over 11,000 R-388/URR receivers were built which seems to confirm the estimated total production. 

All contracts were built by Collins Radio Company although in the early 1950s Hallicrafters supplied a couple of special types of military cabinets along with R-388 receiver manuals that came with the cabinets. These Hallicrafters cabinets and manuals are probably the source of the erroneous "Hallicrafters-built R-388" myth (Collins wouldn't even let RACAL build 51J receivers.) Some (possibly all) of the U.S. Navy R-388/URR receivers were designated as AN/URR-23-A (but were identified as "R-388" on the Navy data plate) and this set-up included the receiver installed in a Collins table cabinet (designated as CY-1235/U) and a Collins 270G-3 speaker (designated as LS-199/U.) Additionally, the USAF had Raytheon supply them with a few modified receivers that were designated AF-30 (R-645) and were used for backscatter ionospheric purposes. These USAF receivers had a blue front panel that lacked the standard grab handles. The largest contract quantities for R-388/URR receivers are from 1951, 1952 and 1953 contracts. Careful examination of the various R-388 data plates will show several different "order" numbers within each year. There were at least seven different order numbers used in 1951 and at least five different order numbers used in 1952.

With the R-388 and 51J-3, the side panels were changed to steel and finished in yellow-cadmium type II plating. Some of the aluminum parts were given a yellow iridite finish but some of the chassis "gold color" is from the MFP coating. All R-388 receivers were MFP coated with special attention to all solder joints being coated. The entire underneath of the chassis was usually coated. MFP, Moisture-Fungus Proofing, was necessary due to the environments where military equipment was used. Grab handles were added to the front panel along with a high quality Burlington Co. sealed meter (some receivers will be found with different makes of meters but they all have similar scaling.) The METER toggle switch was changed to a "spring-return" on the (audio) OUTPUT so that Carrier Level or INPUT was the normal switch position with OUTPUT only reading as long as the switch was held in that position. Two NPO capacitors were added to the BFO (inside the can) to reduce f drift. The conventional "Remote Relay" function that paralleled the front panel "STAND BY" and "ON" functions found on the 51J-1 and 2 was eliminated in the R-388 receiver. Instead a "Break In" function was added that utilized an external +12vdc to actuate an internal relay that disconnected the antenna, grounded the receiver input circuitry and removed the IF plate voltage upon actuation (the IF voltage removal was changed with a 1952 MWO.) The new "Break In" function was entirely separate from the front panel STAND BY switch position meaning that the "Break In" relay could only be actuated using the rear chassis REMOTE terminals. Later, there was also a military supplied modification kit for the end-user addition of a solid-state RF-driven relay circuit to protect the receiver's antenna input circuits from high levels of antenna current induced from nearby transmitters while the receiver was in operation (Break-in off.) This replaced the neon bulb antenna static-drain protection device. These RF-driven circuits were installed in 1957 or later R-388s. Also, in the 1957 or later R-388s will be a Diode Load and AVC test jacks on the rear chassis apron and the receivers will have an IF Gain pot on top of the chassis. Also, the 1957 and 1962 contract receivers weren't MFP coated.  

All R-388 or 51J-3 receivers now had the entire underneath of the chassis protected with a full bottom cover rather than the 51J-1 and J-2's small cover that was just over the receiver's front end. Both the top and bottom covers must remain mounted on the receiver for proper shielding during operation (removed only for servicing or alignments.) The metal bezel used on the J-1 and J-2 was replaced with a black bakelite bezel with no receiver designation on the bezel. Garolite alignment tools, four different size Bristol wrenches and a right-angled Phillip's screwdriver were supplied tools that mounted in special holders on the inner side panel and on the inside and back of the top cover. There are many other changes in component types and placement throughout the R-388 receiver making it quite different from its predecessors, the 51J-1 and J-2. Additionally, throughout R-388 production there were several changes in component values and some circuit changes or additions occurred making the very last two contract versions, the 1957 and 1962 receivers, very different in several minor details from earlier versions.   

A "Collins Manufacturing Serial Number" was punch-stamped (or sometimes ink-stamped) on the rear of the chassis. This number was for Collins "in-house" use since the receivers were built and tested, then put into Collins stock. To fill orders, receivers were pulled from stock, sent through QC and then assigned a serial number that was for the end-user, either the Signal Corps/USN or for civilian users for the 51J-3 receivers. The same serial number that is punch-stamped or ink-stamped on the rear chassis is usually (but not always) white ink-stamped on the front panel but located so that the data plate would cover it. These Collins "in-house" serial numbers won't match the data plate serial number,...and they shouldn't on R-388 receivers.   NOTE: Sometimes an acceptance stamp will be found under the data plate instead of the Collins SN. Also, as the "in-house" numbers approached 10,000 (after 1953,) the numbers returned back to two and three digit numbers. The ink-stamped or punch-stamped Collins in-house SNs appear to be consecutive for each order/contracts but not over the entire production of the R-388 receivers.   >>>

51J-3

The 51J-3 is the most enigmatic of the 51J Series receivers. Most of the information presented in this 51J-3 section is based on examining photographs, reading vintage documentation, hearsay, a lot of assumptions and even more speculation. Some confirmed information was provided by Nick England from navy-radio.com including a photo of a genuine 51J-3. I've never been able to examine a known-genuine 51J-3 "in person" and, of course, that would have been extremely helpful. The final "key identification points" in this section have what seems to be the most logical things to look for when trying to confirm whether a receiver is a 51J-3 or a R-388, if there really was much of a difference. But, as far as trying to logically figure out the "how or why" of Collins production and their internal processes, that's probably impossible.

Is the 51J-3 a R-388? -  The 51J-3 is identical to the R-388/URR receiver except for the following,...

Military Inspection Stamps - The 51J-3, being that it's a civilian-commercial receiver, won't have any military inspection stamps. On the other hand, all R-388 receivers will have military inspection stamps from the Signal Corps or USN located in various places on the panel (sometimes under the data plate) and sometimes on the chassis or the side panels. The presence of mil-stamps is probably the most reliable indicator that can be used to confirm that the receiver is a R-388 and not a 51J-3.

Data Plates - 51J-3 receivers should definitely not have an Army data plate. If a data plate was installed on the 51J-3 it would be the Collins type with the winged emblem at the top of the etched tag with "51J-3" stamped in the TYPE area. This Collins data plate should have machine stamped numbers, not hand stamped with dies. Of the known authentic serial numbers, they are in the 2000 range of numbers but whether these are numbers from a data plate serial number or are actually Collins Mfg serial numbers being used as serial numbers is unknown. Which brings up the question, did all 51J-3 receivers come with a data plate installed? The Collins 51J-3 manual shows the receiver without any data plate mounted (see photo below.) The 51J-3 manual artwork shows the receiver several times without a data plate and it's obvious that the panel isn't even drilled for a data plate mounting (see photo below.) The Howard W. Sams Photofact for the 51J-3 also shows the receiver with no data plate and the Collins Mfg serial number ink-stamp (usually under the data plate) in plain view. I've seen only one photo of a 51J-3 receiver with a Collins data plate that had an appropriately low serial number but, being a photo, there was no way to confirm whether the data plate was authentic or not (and there are a lot of different repros out there.) It's possible that the Collins 51J-3 manual might just be "artwork" and not really representative of the receiver that was supplied to civilian-commercial customers. However, the Sams appears to be a photo since even the Collins Mfg serial number is shown stamped on the front panel. There seems to be sufficient evidence that at least some 51J-3 receivers were sold without data plates and relied on the Collins Mfg ink-stamped number as the serial number.

Collins 51J-3,...or is it?

This R-388 was sold as a 51J-3 because of its repro data plate. A close inspection of this receiver found several USN inspection stamps, a USN ID number that was vibro-etched on the chassis rear apron. However, as far as this receiver's exterior appearance,...this is what most of the later 51J-3 receivers looked like.

PTO Serial Number Prefix or Suffix - Apparently the 51J-3 PTO serial numbers didn't have either a letter prefix or a letter suffix. R-388 receivers will have their PTO serial number shown with a "M" prefix. Nick England sent me a good photo of 51J-3 SN:6636 that was owned by Douglas Aircraft. The serial number is actually the Collins production number and the tag installed was probably made by Douglas (it's a silk-screen tag shown in photo below.) The chassis photo shows the PTO with just a serial number with no suffix or prefix, PTO SN: 945. Note on the cards associated with 51J-3 SN: 2090 show that the PTO is SN: 5991 (no letter prefix or suffix) and this card is dated Jan. 1953. Also, the PTO shown twice in the 51J-3 manual doesn't have a prefix or suffix on the serial number of the PTO (manual dated 1952 but artwork is probably earlier.) While the 51J-3 PTO serial number won't have a letter prefix or suffix, the R-388 receiver PTOs serial numbers always seem to have the "M" prefix.

Were 51J-3s Moisture-Fungus Proof Coated? - The basic assumption would be that civilian-commercial 51J-3 receivers typically wouldn't be MFP treated while all R-388s had to be MFP treated. But, there are several questions about the MFP process at Collins. One question regarding MFP would be when and where the MFP coating was applied on R-388 receivers. It would be common procedure to MFP coat at completion of assembly work on the receiver. However, inspection of the IF and BFO coil assemblies used on R-388s will show that the solder joints inside the shield-can were MFP coated. This seems to indicate that a MFP process was performed at different stages of assembly. Component assemblies would be MFP coated at their assembly and probably put into stock "ready to use." It seems unlikely that IF and BFO assemblies would be stocked differently, that is, IF/BFO assemblies for the R-388 and other non-MFP'd IF/BFO assemblies for the 51J-3. So one could assume that all of these types of assemblies have MFP on their solder joints and were used in either receiver. Since the MFP process was on-going through production, one can assume that some other parts of the 51J-3 were also MFP coated before assembly. And, from there then it's not too difficult to believe that the 51J-3 was MFP coated the same as the R-388. The ink-stamped MFP date is obviously Collins-applied. The 51J-3 manual plainly shows the rear chassis with a MFP stamp dated in 1951. So, 51J-3 MFP coated?,...very likely.

Did the 51J-3 Have a Front Panel Break In Switch? - The 51J-3 didn't have a front panel mounted "Break-in" switch but the 1953 contracts and later R-388 receivers had a "Break-in" toggle switch where the SPEAKER jack had been located.
 

From navy-radio.com , Nick England sent me this picture of one of the few likely genuine 51J-3 receivers to be photographed. There aren't any mil inspections stamps and it was owned by a commercial company, Douglas Aircraft. The receiver was probably shipped to Douglas without a serial number data plate since the one installed is an obvious repro. The serial number stamped on the tag is 6636 and that's also the stamped Collins Mfg serial number on the back of the chassis. It would be interesting to known if "6636" is ink-stamped on the front panel under the data plate. PTO SN is 945 with no prefix or suffix.

Did Collins Know?,...'Cause I don't - One would think that Collins had to know when the receiver was in QC processing whether the receiver was filling a military contract and would be identified as a R-388 or filling a civilian order and would be identified as a 51J-3. How the PTO was identified is the important clue. From observing many R-388s and 51J-4 receivers, it's obvious that ALL R-388 PTO serial numbers have a "M" prefix and ALL 51J-4 PTO serial numbers have a "CR" suffix. Also, all 51J-3 PTOs will not have a "M" prefix.

During R-388 and 51J-3 manufacture, when was the PTO serialized? If the PTO was serialized at calibration that would require Collins to stock "M" and "no prefix/suffix" versions separately. It would require that Collins know at the time of receiver assembly whether the receiver being built was going to be a R-388 or a 51J-3. Then these receivers would have to be stocked separately as either a R-388 only requiring QC to fulfill a contract order. Or, stocked as a 51J-3 only requiring QC to fill a customer order. It all sounds like a process that becomes too convoluted to be practical,...but,...hey, we're talking about Collins.

Logically, a much more believable possibility would be that all 70E-15 PTOs were calibrated and sealed and went into stock that way, that is, "ready to install" but with no label. The receivers were all built with the same type of 70E-15 PTO without a label and stocked as R-388/51J-3 receivers. When pulled from stock and sent to QC, then it was known whether the receiver was going on a R-388 contract or was for a 51J-3 order. During QC, the PTO label was filled out (in ink, by hand) with a serial number from the 70E-15 PTO SN blocks for either R-388 "M" numbers or 51J-3 no prefix or "CR" numbers with the appropriate prefix or suffix added and then the label was applied to the PTO. This scenario makes the testing and stocking of the PTOs much more believable and also allows for having fully functional R-388/51J-3 receivers stocked that could be pulled from stock, then identified properly during QC and finally processed out as either a R-388 or as a 51J-3. Of course, all of this is speculation,...but it does sound like the most believable scenario but that doesn't mean that's the way it happened,...and it probably didn't.
 

Other Evidence Turns Up - Although all of the above sounds logical, that doesn't mean that Collins worked that way. Note the manual shown to the right with the inspection cards stapled inside the front cover. Of particular interest is the "RELEASE TO STORES" green tag in the center. This implies that indeed 51J-3 receivers were stored in Collins stock as complete "ready to go" 51J-3 receivers. So something determined the receiver type at initial assembly when the "non-M" type PTO was installed (already cal'd and tagged probably) and the receiver completed as a 51J-3 and stored as a 51J-3. One other problem exists with the serial numbers of 2090 and 2076,...since these are serial numbers found in 51J-3 manuals that have become separated from their respective receivers, there's no way to confirm whether the numbers shown are off of a data plate or are actually the Collins Mfg serial numbers. It's unknown whether the Collins Mfg. serial numbers were from the same roster that was used for R-388s but since the 51J-3 shown above has 6636 as its Collins Mfg. SN, it seems possible that the same roster was used. If that's the case, then these 1953 era serial numbers of 2090 and 2076 are too low for Collins Mfg. numbers and must be actual serial numbers that were taken from data plates. It's certainly possible that some 51J-3 receivers that were shipped without data plates ended up using the Collins Mfg. serial number, which was conveniently stamped on the front panel right where a data plate would have been located, as the "serial number" of the receiver. As mentioned many times by many people,...it's next to impossible to determine exactly how Collins processed production internally and the relationship between the R-388 and the 51J-3 is about as confusing as it gets. The manual shown to the right is for 51J-3 SN: 2090. It has several of the original inspection cards stamped inside. Dates are from Jan 1953. Note the PTO SN: is 5991            photo from: www.navy-radio.com

No data plate on the Receiver

51J-3 as shown in Sams Photofact - No data plate on Receiver
Collins production number on the receiver is "6574"

From the 51J-3 manual - Not even drill holes for a data plate

51J-3 Key Identification Points

1. The 51J-3 might have a data plate but in many vintage documents the receivers are shown without data plates and may have used the Collins Mfg Number as the serial number. The two known referenced 51J-3 serial numbers are in the 2000 block of numbers and both are from early 1953 but it's unknown if these numbers are off of a data plate or if they are actually Collins Mfg Numbers (the assumption is that they are from data plates.) Data plates, if installed, should be etched tags with machine-stamped ID and serial numbers, not hand stamped with dies. It very possible that some 51J-3 panels that didn't have data plates weren't even drilled for data plates (that's the way the panel is shown in the 51J-3 manual.)

2. The 51J-3 won't have military inspection stamps.

3. The 51J-3 was MFP-coated. The navy-radio.com 51J-3 does have the MFP-coating. The 51J-3 manual shows the rear chassis with a "MFP" stamp. MFP-coating is a "most likely."

4. The known confirmed 51J-3 PTO serial numbers don't have prefixes or suffixes. The PTO shown twice in the 51J-3 manual doesn't have a prefix or a suffix on its serial number. But, the 51J-3 PTO serial number definitely won't have the "M" prefix.

5. Break-in Switch was not used on the 51J-3.

51J-3,...Rare? - So, finding a true 51J-3 will be difficult and will require a lot of searching. Be sure to investigate any supposed 51J-3 carefully. I really think that some of the "tag-less" R-388 receivers, if they were inspected carefully and it was evident that no data plate had ever been installed and, most importantly, NO mil-stamps were found, these receivers would probably end-up being 51J-3 receivers,...especially if said receiver's PTO serial number had no prefix or suffix.

Putting a quantity on the 51J-3 is next to impossible. The two serial number of 2090 and 2076 might imply that some quantity over 2000 51J-3 receivers were produced. That's certainly possible since they were available from 1951 up through 1953. But other questions come up,...like,...if over 2000 51J-3 were produced, where are they? They never show up. I've never run across a 51J-3. I think it's a possibility that many 51J-3 receivers exist but are misidentified as R-388 receivers with "missing data plates.  

So, are the serial numbers 2090 and 2076 actually the Collins production number being used as the serial number? The numbers seem too low to be Collins Mfg SNs for 1953 (Collins Mfg SNs were well-over 6000 in 1952,) so they must be serial numbers from actual data plates. But, since these serial numbers were found written in 51J-3 manuals, there wasn't an actual receiver to verify what serial number was used.

Ultimately, it becomes too confusing because there are so many possibilities with no real physical examples to verify any suppositions. Investigating the 51J-3 possibilities really points to just how little is known about Collins' production process, their product identification system, their stocking methods,...literally everything about Collins' production methods and records keeping. And, when it comes to the 51J-3, nobody knows for sure how Collins stocked, serialized and kept track of things,...they just did.

More Information on Repro 51J-3 tags
Watch Out for 12000+ Serial Numbers

To further confuse this R-388 versus 51J-3 identification problem there are also reproduction 51J-3 tags out there. I received an e-mail from Tom N5OFF regarding these high serial numbers on 51J-3 receivers. About 20+ years ago, Tom obtained permission from Collins to reproduce the 51J-3 and 51J-4 serial number data plates. To assure Collins and Collins collectors that any informed person would be able to easily identify these high quality, authentic but new data plates as "reproductions," Tom stamped the series of 200 tags with serial numbers beginning with 12000 (see photo right.) The key to this, of course, is being an "informed person." The Japanese website adachi-gilken.com has a supposed 51J-3 that has a repro data plate with the serial number of 12047, so the repros are "out there."

The existence of quality repro data plates for the 51J-3 puts into question the identification of any supposed 51J-3 that has one of these repro tags installed. Be very careful when inspecting ANY 51J-3 for intended purchase. Thoroughly check the receiver for military inspection stamps. If Signal Corps stamps are found then the receiver IS a R-388. If the receiver has a Break-in switch, it's a R-388. If there are USN stamps then the receiver is an AN/URR-23-A but would have been tagged as "R-388."  Be suspicious if a supposed 51J-3 has a PTO serial number that has the "M" prefix (double-check for mil-stamps.) And, of course, if the 51J-3 data plate is serialized 12000+ and the other military indicators are also present, then the receiver is a R-388. But, be suspicious of ANY 51J-3 data plate because there are many indications that repro tags other than N5OFF's tags have been made and might be installed. Also, there are many indications that some 51J-3 receivers may have come without any data plate installed (perhaps not even drilled holes for a data plate in the panel.) 51J-3s without a data plate won't have mil-stamps or a "M" PTO.

Always consider that while civilian-commercial 51J-3 receivers do exist they are very scarce, seldom encountered and often misidentified.

- A Reproduction 51J-3 Collins-type Data Plate -
The original 51J-3 data plate, if installed, would have looked like this reproduction but the serial number would have been a number lower than 1500.

In late-1954, the 51J-4, with 19 tubes and three mechanical filters (1.4kc, 3.1kc and 6.0kc actual bandwidth,) became available and was offered up to about 1964. The 51J-4 was to be the ultimate evolution of the 51J Series in some regards. Some users feel that the mechanical filter upgrades seriously degraded the audio reproduction capabilities of the receiver, especially for AM-SW-BC users. On the positive side, Collins had recently developed the mechanical filter to provide an IF passband that was defined with steep skirts and a flat top resulting in superior selectivity. Interference from adjacent frequency signals would just disappear as the IF passband was narrowed as more selective mechanical filters were switched in. As HF band congestion increased, so did the need for a more and more selective receiver. The 51J-4 added a fourth stage of IF amplification to compensate for the insertion loss of the mechanical filters. It will be noted when inspecting a 51J-4 chassis that there is a Mechanical Filter Assembly (Collins 354A-1) that is mounted to the chassis. Under the MF Assembly, the chassis is punched for the R-388 type of IF construction and has the silk-screened nomenclature for the R-388 also. The addition of the MF Assembly Unit includes two 6BA6 tubes mounted on top of the unit that actually amplify the input and output of the selected mechanical filter thus bringing the total number of IF amplifiers to four although there are only three "adjustably-tuned IF amplifiers."

The 51J-4 500kc IF transformers have a 3pf capacitor added from the primary to the secondary to increase the coupling and broaden the IF bandwidth. Similar IF coupling caps, with slightly less capacitance of 2pf, were installed in most R-388 receivers but weren't used at all in the 51J-1 or J-2. The slightly wider IF bandwidth assured the mechanical filters determined the IF selectivity along with the Crystal Filter.   

The mechanical filter selector switch shaft has a chrome lever that is placed behind the BFO knob. The factory 51J-4 receivers will have the MF bandwidth silk screened on the front panel as 1KC, 3KC and 6KC. The Crystal Filter was retained as an additional tool for dealing with interference. Although MFs were excellent for providing a well-defined bandwidth, that factor alone can't eliminate heterodyne interference or enhance certain IF frequencies for better CW reception. Collins knew the combination of MF selectivity along with the Crystal Filter's unique abilities gave the users the best in QRM-fighting tools.

 
51J-4 SN:2392 from 1957 installed in original type cabinet  (owner: KB6SCO)

photo above: The top of the chassis of 51J-4 SN:2392. Note that the chassis is not MFP coated. The Mechanical Filter Assembly just to the right of the PTO. The original plastic bag that's attached to the Crystal Filter top usually contained spare parts, in this case two PL-259 connectors.  NOTE: This 51J-4 now belongs to KB6SCO

 
photo above: Under 51J-4 SN:2392 receiver. On the right is the receiver front-end showing the elaborate 30-postion bandswitch and its multiple sections for the Crystal Oscillator, the Variable-IF and the ANT/RF/Mixer sections with all areas shielded. Visible at the bottom is the neon static-drain bulb, the Antenna Trimmer C and K101 (Break-in Relay.) On the left is the IF and Crystal Filter at the upper left and the power supply chokes. The lower left is the filter cap socket and the AVC, NL and Audio stages.  

As far as the rest of the 51J-4 circuit, there weren't any changes made to the RF front-end of the receiver, the simple envelope detector, the AVC amplifier-rectifier circuit or the noise limiter circuits. The audio output circuitry also remained unchanged. Early in production a yellow plastic-body test jack for the AGC line output and a green plastic-body test jack for the Diode Load output were added to the rear chassis apron. Also, added was a printed notice regarding the FCC Part 15 compliance for the receiver that was applied to the rear chassis apron.

It appears from serial number versus reported details (to navy-radio.com) that the very first few 51J-4 receivers were MFP-coated. This MFP detail only appears in two reported J-4s and they are very early serial numbers. As a general rule, it can be assumed that all 51J-4 receivers won't be MFP-coated.

In the late-fifties, an IF Gain control was added to the circuit. This pot (R187) was located next to the CL Meter Zero Adjust pot on top of the chassis between the MF Assembly and the power transformer. The IF gain adjustment is a 10K pot, cathode to ground adjustment on the 3rd IF amplifier tube. The addition of an IF gain adjustment would allow users to optimize the IF gain versus the noise produced with the end result being better audio with less distortion and lower background noise. The IF Gain circuit is shown in the later edition 51J-4 manuals. Also changed in the very late production receivers was the value of R-155, the cathode resistor for the first AF amplifier. The value was changed from 3.3K to 1.8K in order to increase the available audio output (mentioned as an adendum in the ninth edition manual.) 51J-4 receivers with serial numbers higher than 7000 might be found with S-line style knobs installed (see Swiss Army 51J-4 Receivers further down this page.)

Federal Procurement catalogs show that the 51J-4 was designated as the R-388A/URR or the R-388B/URR but it's doubtful that any receivers were actually identified as such. No "tagged" R-388A receivers have been found which seems to indicate that although the designation was used in catalogs (perhaps for ordering identification) the actual receivers were probably shipped from Collins with a data plate showing "51J-4."

There are reproduction 51J-4 data plates. When N5OFF made the 51J-3 reproduction data plates over twenty years ago, he also made repro data plates for the 51J-4. It's not nearly as much of a concern because there isn't any ambiguity when identifying a 51J-4. The same "high serial number identification" applies for the repro 51J-4 data plates, that is, serial numbers over 12,000 will identify the data plate as a reproduction. I've only seen one 51J-4 for sale that had one of these data plates installed and from the seller's description it was obvious he was unaware that the serial number identified the data plate as a reproduction.

More 51J-4 Receivers

The "Unofficial" R-388A/URR

  Collins 354A-1 Mechanical Filter Kit Installation in a 1951 R-388/URR Mil SN:33 Collins SN:5282
 

The R-388A/URR designation is found in the Federal Procurement Books as a purchasable Collins receiver with mechanical filters priced at $1200. No known example exists and the assumption is that when the R-388A was ordered, Collins would send a 51J-4 and the receiver's data plate (if installed) was the standard Collins tag with 51J-4 shown as the TYPE of receiver. It's become somewhat common for Collins collectors to casually refer to R-388 receivers that have had the 354A-1 Mechanical Filter Kit installed as a "R-388A" receiver. This is an informal and unofficial reference but since no know example exists of an actual R-388A receiver, maybe these conversions are as close as Collins collectors can come the non-existent "real thing."

The 354A-1 Kit - Shortly after the 51J-4 became available, Collins offered the 354A-1 Mechanical Filter Kit. This was exactly the same assembly that was installed in all 51J-4 receivers. A close look under the chassis of any 51J-4 and one can see that the chassis holes are punched and the nomenclature is silk-screened for the assembly of 51J-3 or R-388 receivers. So, essentially the 354A-1 kit allowed the owner of a 51J-3 to upgrade that receiver to a 51J-4. At the time, owners of R-388 receivers would have been the military and there's quite a bit of evidence that some 354A-1 installations into R-388 receivers were military depot jobs. It was also possible to upgrade a late-version 51J-2 into a receiver that had mechanical filters for improved selectivity. The 51J-2 serial number had to be 501 or higher. The 51J-2 chassis layout before SN:501 didn't provide enough space for the 354A-1 installation.

The 354A-1 kit installation wasn't for beginners. Its successful installation required an electronics technician background and experience. The 28 page instruction manual contained 78 numerated steps (with check-off boxes) to guide the installer. The instructions also contained a parts list, schematic, drawings of the installation, lists of many components and parts that needed removal and finally, IF alignment instructions. Besides an electronics background and experience, the installer would have required a good mechanical aptitude for the correct assembly so that everything would function correctly, not only electronically, but mechanically. An excellent soldering technique, real SnPb solder, quality tools all were necessary for the end result to work as designed.   

Several components had to be removed, including the first IF transformer, the 1st IF amplifier tube socket, many resistors and capacitors. There were two small 3pf coupling capacitors that had to be installed across the 2nd and 3rd IF transformers to widen the IF bandwidth to about 8kc. This was to allow the 6kc mechanical filter to limit the actual IF bandwidth. It was imperative that the insulated BFO shaft coupler (a supplied part) be used otherwise noisy BFO operation would result.

1951 R-388 with the 354A-1 Mechanical Filter Kit Installed

The "Unofficial R-388A" is actually a 1951 R-388/URR with a 354A-1 Mechanical Filter Kit installed
 

Mechanically, the position of the MF kit on the chassis, the alignment of the BFO shaft and the MF switch shaft through the .250" bushing and the mounted position of the BFO assembly all had to be adjusted into physical alignment in order for the BFO operation to be smooth and for the switching of the MF selector to operate properly.

The installation of the 354A-1 into a R-388 receiver was further complicated by the MFP coating of all solder joints and most components. Reworking electronic devices that have been MFP coated requires special training and special tools for cleaning off the old MFP from the solder joint first. Then the joint is unsoldered using proper solder removal tools, the component or wire removed and then the joint is further cleaned of excess solder. If another wire is to be installed its end should be reconditioned and tinned. A new wire would be stripped and tinned. If it's a new component to be installed then the leads should be clean and straight. The installation requires a full wrap, then the joint is soldered, then another cleaning to remove the solder rosin residue and, finally a coating of MFP applied with a small paint brush just to the reworked joint. It's a difficult process but professional rework will barely be detectable while amateur rework can do catastrophic damage.

The Unofficial R-388A receiver shown is a 1951 R-388/URR with Order 3096-PHILA-51 and SN:33 on its data plate and the Collins manufacturing SN:5282 punch-stamped on the rear chassis. Also, the receiver has a MFP date of "OCT 1952" stamped on the rear chassis. Besides the 354A-1 MF Kit installation, there's also a newer 70E-15 "replacement" PTO installed. This can be determined by the PTO serial number "11277," indicating this PTO was probably built in the mid-1960s (after production of both the R-388 and the 51J-4 ended.)

This R-388 conversion using the 354A-1 MF Kit was probably one of the worst examples of careless, incompetent, destructive and worse-than-amateur level workmanship I've come across in quite a while. The refurbishment and repair work required was extensive. If you like to read about that sort of thing and sort of vicariously commiserate with me while I sorted things out, the 1951 R-388 with 354A-1 Mechanical Filter Kit installation is written-up in my typical OCD excruciating detail style in Part 3 of this web-article (link to Part 3 is at the bottom of this page.)

Are 51J-4 Receivers with Light-Gray
 Front Panels Laboratory Receivers?

Some 51J-4 receivers will have light-gray panels with black nomenclature. It's generally believed that these were laboratory or test facility receivers. Some sources say the light-gray panels were used at Collins in-house laboratories and test set-ups.

Some light-gray panel J-4s were used in Beckman test equipment. It's likely that Beckman produced the panels since the J-4 receivers also required some Beckman-implemented circuitry modification to function as necessary for the Beckman equipment. The Beckman panels will have special nomenclature that included Beckman or Beckman/Berkeley logos, the panels aren't drilled for a Collins data plate and also sometimes elimination of the CALIBRATE function and its nomenclature was necessary. All of the light-gray panel Beckman receivers seem to be marked as Model 7700 with various suffixes to distinguish some of the circuit modifications required for their various uses. It seems likely that Beckman bought standard 51J-4 receivers and performed all of the modification work themselves.

The other 51J-4 receivers with light-gray panels are standard Collins receivers with the exception of the panel. These panels are drilled for data plates which are sometimes installed. Some examples will have a mixed variety of Collins-type knobs. There are two types of nomenclature fonts seen on the light-gray panel 51J-4 receivers. The standard is the small letter-number font that's common to both the St. James gray and the light-gray panels but also seen is a larger letter-number font as seen on receivers #2 and #3 shown further below.

51J-4 SN:4723 from 1960

Note the asset number shown on the bar code tag. This receiver has the asset sn:146 while my receiver has the asset sn:00142. Note the Collins-Rockwell sticker upper left. Data plate removal is a common occurrence with the lab receivers.    photo from: Universal Radio

Light-Gray Panel 51J-4  sn: 4723

Making the Case for a Collins Lab Receiver

From some of the features found on this receiver one might infer that it was used at Collins. For example, it has a Collins-built plug-in RC coupling adapter in place of the 6kc mechanical filter. Also, the 1.4kc MF was replaced with an 800hz MF (J-4 type.) But wait,... here's more definitive evidence. 

I suspect that the number "029 1059 00142" engraved at the upper right corner of the panel on SN:4723 is an asset number that identifies the receiver as property of a particular company or laboratory. By examining other light-gray panel 51J-4 receivers, it seems that "029 1059" is the Collins' asset number with "00142" indicating the particular receiver by an "asset serial number." Three other receivers are shown with identical "029 1059" numbers and varying serial numbers. The website www.navy-radio.com  has a photo of a light-gray panel 51J-4 receiver with the asset number is "029 1059 00" which, if the serial number of 105 was added after the "00," becoming "029 1059 00105," then the format would be the same and with nearly the same numbers that SN:4723 has. Does that imply that 4723's asset number identifies the receiver as sn:142 (for the asset records anyway?) Shown to the left is a photo from Universal Radio of a light-gray panel 51J-4 they sold. The sticker shows basically the same format for the asset number, dropping the "00" ahead of the asset serial number, and a Collins-Rockwell sticker. Both sn 146 and sn 105 don't have data plates installed. Missing data plates are common on light-gray panel receivers and may have been part of the deposition of the receiver or maybe they were never installed.

Close-up of the light-gray panel 51J-4 from www.navy-radio.com with same engraved asset number but with the serial number separated. Note the toggle-boot

The third receiver is shown below in the collection of light-gray panel 51J-4 receiver photos. Receiver #3 has the same style engraved asset number in a similar location. This receiver appears to still have its data plate. Interestingly, all receivers shown seem to have the same style toggle boot installed. Unfortunately the poor resolution of the photograph prevents reading the asset number.

The most significant "Collins-clue" comes from the "Collins - Product Support Operations" sticker. The big question is, what does the sticker indicate? "Product Support Operations" does sound like an engineering lab. "Processed" also sounds like an "in house" thing,...but, then,...why bother with the sticker? Unless the sticker indicates which division of Collins the receiver belonged to. Center photo below shows the Collins sticker on the rear chassis of sn:4723. Also, of note is the photo of "029 1059 146" above with its Collins-Rockwell sticker.

Also of interest on sn:4723 is the single digit "production" serial number "2" that's ink-stamped on the back of the chassis - below left. Note the FCC Part 15 compliance stamp. And, below to the far right, a close-up of the data plate.

Other Light-Gray Panel 51J-4 Receivers

The photos below show a few other variations of the 51J-4 receiver with light-gray smooth-finish painted panel. Note that most have the standard Daka-Ware knobs installed. Receiver #2 has S-line knobs installed along with the standard Daka-Ware tuning and band change knobs.

The Beckman Model 7700/3 shown in photo #1 is very different from the Beckman/Berkeley system receiver shown further below. Beckman #1 appears from the front to be almost a standard J-4 receiver. But, who knows what's behind the front panel? Both Beckman 7700s use slotted OH screws to mount the dial bezel and the "Berkeley" uses all slotted OHs on the panel.

Note the engraved asset number above the CL meter on #3 receiver. Note the larger nomenclature on #2 and #3 receivers. I wonder when the 4:1 vernier tuning knobs were added to #1 and #3 receivers.

Receiver #4 is an unusual Beckman 7700/3 that's under restoration. The receiver was found by Tom W3TA in nearly complete condition but seriously suffering from "outdoor over-exposure." This photo shows the very different greenish-gray wrinkle finish Beckman panel after its clean-up.

The Beckman Frequency Measuring System photo and photos #1, #2 and #3 are from Bear - WB2GCR    Photo #4 is from Tom - W3TA

1

2

3

4

Beckman/Berkeley Frequency Measuring System

Description - Beckman/Berkeley supplied these Frequency Measuring Systems to allow accurate measurement of any RF signal in the frequency range of the 51J-4 receiver, or .5mc up to 30.5mc. Signal levels could be as low as 1uv and they could still be measured. There were three basic units, the 51J-4 receiver, the Beckman Translator that converted signals from the 51J-4 into various types of signal information to interface with the Beckman Digital Frequency Counter, referred to as an "events per unit-time meter." In the rack shown to the left, the Frequency Counter/Meter is at the top. The next device is a separate Frequency Meter. Next down appears to be a power supply. Next is the 51J-4 receiver and below it is the Beckman/Berkeley Translator. This set-up is somewhat different and larger than the smaller "three unit" type of Frequency Measuring System.

Modifications - The 51J-4 did have some modifications made to it. First was to bring out the PTO signal to a connector on the rear chassis of the receiver. Also, an output for the 2nd IF was brought out to the rear of the chassis. The 51J-4 was powered up with a connection to the Frequency Counter/Meter. The 51J-4 shown in the photo left doesn't have a 100kc calibrator installed so there's no switch for it on the front panel (a calibrator would be superfluous on a frequency measuring device.) Also, there's a vernier frequency control to the right of the RF Gain control. It's difficult to read the dymo-lables for the mechanical filter selector but the 1kc is covered with a blue dymo-lable that appears to have "0" on it. The 3kc is a green label and appears to have 7 on it and the 6kc is also a green label that appears to have "9" on it.

Martin also has a modified R-388 that was used by the Swiss Army that has the 354A-1 mechanical filter installation. He also has an earlier version 51J-4 that was #4 in the Collins receivers purchased by the Swiss Army. SN on that J-4 is 1796. The first two 51J receivers, #1 and #2, were listed as 51J-3 receivers and the remaining 82 purchases were 51J-4 receivers. Each receiver was installed in its own wooden case (that was also numbered.) Again, because these Collins receivers were always professionally maintained by the Swiss Army Signal's Shop, they are in excellent condition. Martin reports that SN: 1796 "works as expected."

I've noticed that Martin's photo of his 51J-4 is very popular on the Internet, but he did send me this photo/link via an e-mail plus he included all of the history of these later 51J-4s as used by the Swiss Army.

* This paraphrased quote is from a Technical Materiel Corporation's GPR-90 Sales Service Information (Sup-1 GPR 9-56) and it expresses the frustration that TMC had with users comparing the vague frequency readout  of the GPR-90 to a precise Collins receiver frequency readout. The complete quote was "People who require frequency meters should buy frequency meters - not make them double as communications receivers."

General Information on the Various Types of 51J Receivers

51J-1 - 1947? but probably 1949 through early-1950 - Officially available in 1949 but perhaps supplied to the Signal Corps and a few other commercial users somewhat earlier. Collins (well,...the CCA) says the receivers were "available in 1945" but, if true, this must have been in prototype-form when the design was nearing completion. The basic electronic design concept is certainly obvious in the 1947 75A-1 ham receiver, so it might be possible that the general coverage 51J was concurrently available with the 75A-1 although the sparse quantity of 51J-1 receivers seen today indicates that very few were produced (usually the 51J-2 is seen much more often.) Most sources indicate 1949 as the date of availability to the civilian commercial market and this is when advertising first appears and probably when most of the production took place. When a complete and original 51J model is found it will be equipped with an illuminated amber scale S-meter, the Collins "winged emblem", a metal dial bezel painted satin-black with "51J RECEIVER" silk-screened in white above the kilocycle dial, the 70E-7A PTO, the megacycle dial featured green highlighted amateur bands, "100 KC CRYSTAL" nomenclature was used for the Calibration Oscillator switch only on the prototypes, "CALIBRATE" replaced "100 KC CRYSTAL" on panel nomenclature on all production receivers, the bottom of chassis had a small cover for the RF sections only with the rest of the chassis bottom uncovered, no grab handles were used, no skirt was on Megacycle Change knob but it did feature a crank to ease band changing, a single phone jack on front panel and the circuit used 16 tubes. The antenna input impedance was fixed at 300 ohms nominal Z with no Antenna Trim control used. Each Antenna coil had a primary winding on the J-1 and J-2 receivers. Production level was very low, estimates are less than 1000 receivers were built (including R-381/URR versions.) Unfortunately, of the surviving 51J-1 receivers I've seen, ALL examples had at least several modifications and most had many extensive modifications. Unfortunately, almost all of the 51J-1 examples I've seen had been destructively modified in past attempts to "upgrade" the receiver and usually the destruction was beyond any restoration.

51J-2 - 1950 - The 51J-2 models had essentially the same characteristics as the J-1 with these following exceptions; the meter was changed to a Carrier Level/Audio Output Level meter with appropriate scaling but meter was still the square front bakelite housing model with illumination. A switch was added to the meter circuit to allow measuring either carrier level or audio output, marked "METER - INPUT - OUTPUT." Beginning with 51J-2 serial number 501, the IF transformers were moved further back towards the rear of the chassis. This required moving the three 12AX7 tubes to the left (viewed from the front) and moving the 6AQ5 tube next to the 5V4 rectifier. All subsequent 51J receivers have this same tube-chassis layout. It might be possible that the green highlighted amateur bands were removed from the megacycle drum dial but this is more likely the result of a recent replacement (R-388) drum being installed. Production level of 51J-2 receivers was well over 2000 receivers (2233 highest serial number seen.) Very few of the 51J-2 receivers have survived in totally original condition. All examples found seem to be modified in some manner but generally not as extensively as the J-1.

R-388/URR - late-1950 through 1962 - This is the military version 51J receiver that featured major changes to the original 51J receiver design. The antenna input impedance changed from fixed 300 Z ohms to low-Z 50 ohms adjustable by eliminating the primary winding on the antenna coils and adding an antenna trimmer capacitor with front panel control. The dial bezel was changed to a black bakelite piece and "51J RECEIVER" nomenclature eliminated, the megacycle dial drum no longer had the green highlighted amateur bands, the carrier level meter was now a non-illuminated, sealed unit made by Burlington Company (field replacement meters made by Marion Electric and other companies, DeJur and Simpson are often found as original installations,) audio outputs on the front panel allowed for phone (4Z) or speaker (600Z) connections (pre-1953,) grab handles were now installed on the front panel, a skirted-knob used for megacycle change, an 0A2 voltage regulator tube was installed, the entire bottom of receiver chassis was now protected with a slide-in aluminum bottom cover, the side panels are made of steel and finished with gold color cadmium-based chemical treatment, the schematic was usually applied to the inside of the top cover, all receivers were MFP-coated (very late contract versions were not MFP'd - 1957 and 1962 contracts,) remote standby or Break-in now required +12vdc to be applied to a relay via rear terminals. All R-388 receivers used the new but problematic 70E-15 PTO which added the PTO buffer output tube. This PTO was cylindrical in shape and has two tubes. All R-388 PTOs are identified with a "M" as a prefix for the serial number. An SO-239 connector on rear chassis, "IF OUTPUT," was provided for driving RTTY TUs and other data devices. 1953 and later receivers have a "Break-in" on-off switch added to the front panel in the same location as the SPEAKER jack which was removed. Diode Load and AVC pin jacks are added to the rear chassis on last of production (1957 and 1962 contracts.) The R-388 was supplied to Army Signal Corps at various times from 1950 up into the early 1960s. Most (all?) USN R-388 receivers are considered as AN/URR-23-A but data plate has "R-388" as the receiver ID (on a USN data plate.) At least 11,000 R-388s were produced (as indicated by PTO serial numbers.) Finding all-original, complete and unmodified R-388 receivers is becoming more and more of a rare occurrence. Many of the R-388s found today suffer from incompetent workmanship that seems to have been perpetrated ever since the receivers became easily available on the surplus market and from other sources. But, even now in 2023, finding an "all original, unmolested" example is still not impossibly difficult.

51J-3 - 1951 through 1954 - The 51J-3 is virtually identical to the R-388/URR but the ID tag will show that the receiver is a 51J-3 rather than R-388/URR (there is increasing evidence that many 51J-3 receivers were sold without ID tags and used the Collins production ink-stamped number on the front panel as the serial number.) As far as the circuit and the components used, the civilian 51J-3 receivers are exactly the same as the R-388/URR. Even the 51J-3 chassis was MFP coated. But, the 51J-3 receiver's PTO will have a serial number without a letter prefix or suffix, just a number. It could be possible that the last of 51J-3 production receivers might have a "CR" suffix on the PTO serial number. 51J-3 receivers are sometimes installed in a Collins cabinet. 51J-3 receivers wouldn't have any military inspection stamps. Check for Signal Corps acceptance stamps or other mil-stamps for confirmation. Navy stamps indicate the receiver is the AN/URR-23-A but it would have been tagged as "R-388" on a Navy data plate. 51J-3 serial numbers in the 12000 range indicate that the data plate is a reproduction and calls into question the authenticity of the receiver as a true 51J-3. While many "radiophiles" consider the R-388 is a 51J-3 (and they very nearly are,) the subtle differences are very apparent to Collins enthusiasts. Authentic civilian 51J-3 receivers are scarce. But, the 51J-3 receivers that were sent out without data plates had no overt identification (actually, no identification at all) so it's possible that many 51J-3 receivers might be misidentified as a "tag-less R-388" receiver.

51J-4 - 1954 through 1964 - Civilian, Commercial, Military versions are all the same with the receiver itself being very similar to the R-388 with the following exceptions. The 51J-4 added three mechanical filters to the IF by installing the Collins 354A-1 Mechanical Filter Kit into the R-388/51J-3 chassis. Examine the 51J-4 chassis and it will be seen that the chassis is punched for the standard IF strip of the R-388 but that a mechanical filter assembly package has been installed which converts the receiver to a 51J-4. The front panels were also changed to add 1KC, 3KC and 6KC nomenclature for the mechanical filter selector switch lever which is located behind the BFO knob. The MF assembly adds an additional IF amplifier tube bringing the tube total to 19. All 51J-4 receivers have a 70E-15 PTOs with "CR" as a suffix for the PTO serial number. Majority of production receivers have St. James Gray wrinkle finish front panels with white silk-screened nomenclature. A small quantity of 51J-4 receivers were produced with front panels painted light-gray smooth finish with black nomenclature, possibly built for laboratory or test facility users. Some 51J-4 light-gray panel receivers are marked "Beckman - Receiver Model 7700" or sometimes "Beckman/Berkeley." The Beckman/Berkeley receivers were a component of a Frequency Measuring System that was a rack with several other components that functioned with the Model 7700/51J-4. The light-gray panel 51J-4 receivers are sometimes found with a variety of Collins-type knobs installed. 51J-4 receivers are not usually MFP coated. AVC and Diode Load pin jacks are on the rear chassis. IF Gain control is on later receivers and is mounted on upper chassis between MF assembly and power transformer. "FCC Part 15" information is silk-screened on the rear of the chassis on later production receivers. When supplied to the military as the R-388A/URR or R-388B/URR it's possible that the ID plate had "51J-4" stamped on it. Serial numbers go up to about 7500. Good examples of "all original and complete" 51J-4 receivers are still very easy to find but, of course, "good ones" will be expensive. There are several J-4 variations that might be considered rare or scarce when compared to the standard J-4.

R-388A/URR and R-388B/URR - With this designation, these receivers should have been Signal Corps receivers. These are versions of the 51J-4 but it's unknown if the data plate actually identified the receiver as a "R-388A/URR." Same with the R-388B. At this time, no examples of a receiver identified by the data plate as "R-388A/URR" are known to exist (navy-radio.com also indicates the lack of any known examples.) The designation is shown as "R-388A/URR" in Federal Procurement Books. It could be that this was just an identification for ordering the 51J-4 receiver and the receivers shipped from Collins may have actually had the "51J-4" designation on the ID tag (or maybe no data plate at all.) Since the R-388A/URR was shown in Federal Procurement Books as a purchasable receiver (purchase price was about $1200,) the implication would be that a military depot conversion of a standard R-388 receiver using the 354A-1 Mechanical Filter Kit making the receiver into a 51J-4 really wasn't a "R-388A,"...or was it? The known examples of mil-converted R-388s to 51J-4 all have their original R-388 data plates still installed. But, after the mil-conversion, how would the mil-users have identified the receiver since it wasn't a standard R-388 anymore?

AN/URR-23 and AN/URR-23-A - These are U. S. Navy designations for the 51J-2 and 51J-3 installed in a Collins cabinet designated as CY-1235/U with Collins speaker designated as LS-199/U. According to the USN manual for the AN/URR-23-A, the receiver IS a R-388 and this is shown many times throughout the manual. Also, three artwork pictures show the data plate which identifies the receiver as "R-388" only, no "/URR" and the data plate has a USN contract number to the lower left of the tag. It's possible that the AN/URR-23-A identification was on the data plate that was mounted on the CY-1235/U cabinet. The LS-199/U loudspeaker also had a data plate installed. The contract was issued for the AN/URR-23-A on June 22, 1951.

R-381/URR and R-381A/URR - These are Signal Corps designations for the 51J-1 and 51J-2. These receivers don't have the Collins Winged Emblem but have typical mil-style data plate installed.

AF-30 or R-645 - Special modified version of the R-388/URR used for ionospheric backscatter research. Modified by Raytheon for the USAF or USMC. Has a blue panel and no grab handles. Supposedly a very small contact of only seven receivers. Several photos on navy-radio.com

Light-Gray Panel 51J-4 - The most often heard reason for the light-gray panel 51J-4 were that they were for use in laboratories or in test facilities. Some say that they were for Collins "in-house" labs. But, did Collins  supply light-gray panels to other end-user laboratories? Were light-gray panels ever advertised as an option?

Receiver Model 7700 - This identification is on the Beckman-versions of the 51J-4 with light-gray, smooth-finish painted front panel. One Beckman/Berkeley 51J-4 has been found with a greenish-gray wrinkle finish panel with identification as Model 7700/3. Various "/xx" suffix to most Model IDs, e.g., Model 7700/2-1. Some of the Model 7700 receivers were components of a Beckman Frequency Measuring System. The Model 7700 receivers had several modifications incorporated into the circuitry so they could function as needed with the other Beckman equipment. Beckman did all of the modifications to the receivers and probably made the panels themselves. It seems probable that Beckman purchased standard 51J-4 receivers from Collins and then incorporated all of the modifications themselves.

Gloss-Black Panel 51J-4 - Two are shown on navy-radio.com. One is in good shape having been restored, the other is in rough condition with modifications. Another original condition one showed up on an Antique Radio Internet forum. The black panel was a replacement panel in black glossy finish and the top and bottom corners of the panel were notched and the grab handles were mounted low on the panel and had ejector buttons. There are no phone jacks on the front panel. There are two fuse mounts above the Crystal Filter controls. The nomenclature for the Crystal Filter was relocated to below the controls. There is a green indicator lamp by the "ON" nomenclature of the power switch. Speculation was that it was the forerunner to the LTV-G133F, a modified 51S-1, for airborne intelligence use. Several photos on navy-radio.com   NOTE: Mar 2023 - One of these variations of the 51J-4 has been found with a black wrinkle finish front panel. The receiver appears to be a R-388 with similar modifications that the gloss black 51J-4 versions had but without a 354A-1 mechanical filter kit. This receiver has been modified over the years and many parts are missing so authenticity is difficult to establish (just examining photographs.)
 

R-648/ARR-41 - The Airborne R-390A/51J-4

While the popular nickname "Airborne R-390A" is certainly appropriate for the R-648, much of the circuit design borrows heavily from the Collins' 51J series of receivers, especially the dual variable "odd-even" IF with reduced number of crystals in the Crystal Oscillator and the fixed 500kc IF. Even the 500kc IF Mechanical Filters are physically a very close match to those found in the 51J-4 (except they're soldered in place.) The R-648 is more like an airborne 51J-4 with modular construction and a mechanical-digital frequency readout.

The R-648 was for use in Navy aircraft (doing RTTY in some airborne installations.) Seventeen tubes are used in the receiver providing two RF amplifiers, double conversion for most bands, three 500kc fixed-frequency IF amplifiers with two mechanical filters (1.4kc and 9.4kc, later models have 1.4kc and 6.0kc MFs) and an audio amplifier with three stages of amplification. When looking at the chassis, one sees the familiar slug racks and slugs, a PTO, modular construction with seven modules and, of course, a mechanical-digital frequency dial. One feature the R-648 has that neither the R-390A or 51J-4 has is tuning from 190kc up to 550kc. The remaining frequency coverage is 2.0mc up to 25.0mc. The R-648 shown is from 1957.

On board the aircraft, the power was usually +28vdc that was provided by the battery-charger buss. This powers the R-648 via an onboard dynamotor with an output of +250vdc at 100mA. The +250vdc is also routed to an 0A2 regulator tube to provide +150vdc. Additionally, a divider network provides about +31vdc for AVC bias. Tube heaters are wired in series-parallel to run on +28vdc and the dial lamps are in parallel on the +28vdc line in the receiver (#327 lamps.)

The GAIN control functions as the RF gain when the receiver is in the CW mode with the AF gain automatically set to maximum. The GAIN control functions as an AF Gain control when the receiver is the VOICE mode with the RF gain controlled by the AVC line. The Sensitivity control is provided to set the maximum available sensitivity and is a slotted-shaft pot behind the "toilet seat" cover marked SENS ADJ." The audio output level is set with a pot adjustment located at the rear of the chassis. This setting was to act as the "maximum" limit so that the headset used would not be over-driven. The audio output impedance is not critical and anything over 300Z ohms was considered appropriate. The audio output has ample volume if a 600Z ohm load is provided. This can be a loudspeaker with a 600Z ohm matching transformer.

In actual use the R-648 is a very sensitive receiver that provides an accurate frequency readout and excellent stability. The selectivity is steep-sided as expected with mechanical filters but the early receivers had a 9.4kc MF for VOICE reception and many times that's just a little too wide in a crowded ham band. The audio filter can be switched in but it's easier and more effective in the AM mode to just tune up or down a couple of kc to reduce QRM. CW is fine with the 1.4kc filter and SSB also sounds just fine with the narrow bandwidth. The CW audio filter is very narrow. There is some vibration from the dynamotor and although it can be felt it doesn't seem to affect the receiver in any way. Noise from the dynamotor might seem a loud at first but, after the audio comes up, the dynamotor noise is pretty much masked by the receiver audio. A nice performing, small, light-weight receiver that's easy to find room for in any vintage ham shack. The shock mount shown isn't original but it does the job of elevating the receiver off of the bench and providing vibration isolation via its shock feet.

51J Cabinets

I've had two 51J-4 receivers that were in what appeared to be their original cabinets but, with both receivers, the top cover couldn't be removed when the receiver was installed in the cabinet because the rear cut-out height wasn't above the second row of side vents and the cutout's width wasn't >17.5". If you can't remove the top cover out the back opening, then suspect an amateur-modified cabinet. To remove the 51J top cover with the receiver mounted in an original Collins 51J cabinet,...lift the cabinet lid to be able to see the wing nuts, then reach around the sides of the cabinet to loosen the wing nuts, then lift the rear part of the top cover up to clear the wing nuts, then slide the top cover back and down and out the rear cabinet opening.

51J Cabinet Artwork
from the 1959 Collins Catalog

Later versions of this cabinet, as found on some 51J-4 receivers and probably on those cabinets sold separately by Collins, had a top lid latch. Also, the tapped panel mounting holes were replaced with drilled holes and clip-nuts. The top lid hinge was changed to a "piano hinge" so it's visible with the lid closed. Rubber feet mounted with aluminum spacers were used on these later receiver cabinets.

1952 R-388/URR - Order:3362-PHILA-52 - Mil SN:161 - Collins SN:9108
installed in a Hallicrafters CY-1260/G Receiver Case

More details on R-388 SN:161 and on the CY-1260/G case in the restoration section in Part 2 of this web-article.

Military Cabinets aka CASE, RECEIVER - Hallicrafters supplied a CY-1260/G cabinet, shown in the photo to the left, that was used with the R-388/URR in some applications. Hallicrafters also supplied a manual when the receiver was installed in this or similar Hallicrafters' cabinets. Also, there was another Hallicrafters cabinet made in a similar style to the CY-1260/G that would hold two "stacked" R-388 receivers. Hallicrafters actually made several CY-version cabinets that were available for several different types of receivers including the SP-600 and the R-274 and R-274D. These Hallicrafters-built cabinets and the accompanying manuals have probably led to the myth that Hallicrafters built R-388 receivers which, of course, wasn't the case. Collins wouldn't even let RACAL build 51J receivers!

The Hallicrafters-built military cabinets are fabricated with heavy steel and welded seams. Shock-type (rubber cushioned) feet are bolted to the bottom of the cabinet and the feet centers are drilled and tapped for secure mounting to a table. There's a rectangular opening in the cabinet-bottom that must have been for ease of manufacture since it can't be for access underneath the receiver because the receiver's bottom cover must remain installed during operation. Inside, an elaborate guide/glide set-up was installed on each inner-wall to mate with accessory metal rolling wheels that are mounted on each of the rear sides of the R-388 receiver. This allowed very easy installation and removal of a receiver from the cabinet since sliding friction was practically eliminated. The cabinet's rear cutout allows easy access to the receiver's rear terminal strips and SO-239 fittings. Six small louver vents are on each side of the cabinet but no grab handles or slots were provided which makes moving this very heavy (~35 lbs. empty!) cabinet somewhat difficult. Note in the photo to the left that the receiver is raised off the table by about two inches when using the CY-1260 cabinet. Total cabinet height is 13.5 inches. Contract date for CY-1260/G cabinets and manuals was in 1951.
 

photo left: 1952 R-388 SN:161 - installed in the Hallicrafters CY-1260/G military cabinet. These cabinets are built from heavy-gauge steel with elaborate glide rails on each inner side, also made of steel. There aren't any grab slots on the sides so moving these cabinets can be difficult with the receiver installed (total weight is 70+lbs.)   photo right: Close-up of the data plate on top of the cabinet.

How Collins Radio Company and the 51J were Responsible for the RACAL RA-17

Things changed for RACAL in 1953 when the British Royal Navy became interested in having RACAL build a couple hundred Collins 51J receivers for Royal Navy use. The 51J was being successfully used for RTTY communications in the USA military. It provided frequency-solid stability along with a dial readout that was "to the kilocycle" accurate. The 51J was the first receiver used by the US military that provided reliable RTTY reception without modifications. The US military R-388/URR was the most successful 51J version and was produced in the greatest quantity. Certainly the Royal Navy was aware of the success the 51J/R-388 afforded US military RTTY users and they needed the same type of reliability for their RATT comms (British for RTTY.)

If RACAL was going to become involved in British-built Collins 51J receivers, they would have to be licensed by Collins to do so. Naturally, RACAL wanted to use mostly British parts (probably so did the Royal Navy) but Collins Radio Company insisted that only parts from the USA could be used and only assembly of Collins 51J receivers could be performed by RACAL. Ultimately, the proposal ended up with a group from Collins Radio doing an inspection of the (then) small RACAL manufacturing facility. Apparently, Collins wasn't impressed and simply refused to license the manufacture of the 51J by RACAL. The only option left was for RACAL to design their own receiver that would meet the requirements for the Royal Navy contract.

RACAL contacted Dr. Trevor Wadley, a well-known South African electronics engineer, to design the receiver and head the project. The result was the famous RA-17 receiver. Though entirely different in electronic design concept, using the famous "Wadley Loop" to eliminate drift, the RA-17 met or exceeded the 51J specifications and went on to be produced in many variations over the next 20 years. In looking at the RA-17 receiver to the right, it's immediately apparent that the physical appearance of the RA-17, especially the dual-dial arrangement and the slanted top (can't be seen in the photo,) paid homage to the Collins 51J receivers. 

RACAL RA-17C-12 from 1961. The "C" versions of the RA-17 were built in England but were intended for North American users and for NATO. Mostly USA-types of tubes, hardware and nomenclature were used for the "C" versions. The RA-17C-12 versions were also designated as the R-501/URR by the USA mil-users. The similarity of the dial escutcheon that uses a slide-rule dial over a circular dial and then the use of a slanted rear part of the top cover all paid homage to the Collins 51J.

          CONTINUE to Part 2                         GO to Part 3                              RETURN to Home Index