Radio Boulevard
Western Historic Radio Museum

Rebuilding the Hammarlund SP-600

~ General Information about the SP-600 ~

~ The Molded Capacitor Problem ~

~ Receiver Disassembly and Capacitor Replacement ~

~ Electronic Rebuilding and Mechanical Restoration ~

 ~ Fixing the Chronic Dial Slippage Problem ~

~ RF/IF Alignment ~

~ Performance Expectations ~

by: Henry Rogers - WA7YBS - WHRM
 

The 1953 Hammarlund SP-600 JX-21

PART 1

Rebuilding the Hammarlund SP-600 Series of Receivers

General Information about the SP-600

Hammarlund released an advertisement in 1948 that introduced their newest receiver, the SPC-600X. The receiver was to sell for $395 but, as Hammarlund had done before, this advertisement was really just to let hams and interested commercial users know what was going to be available in the future. No SPC-600X receivers were ever created other than maybe a prototype or two. It was well-over a year later, in 1950, that Hammarlund introduced the SP-600 Series and the selling price had dramatically risen to nearly $1000. This new SP-600 was primarily intended for the military and commercial user market. It was a very popular receiver and many thousands were built, especially for military applications. It's known that much of the SP-600 design criteria actually came from the U.S. Army Signal Corps, especially the JX crystal oscillator concept and the turret band switching. The "JX" crystal oscillator wasn't new for the SP-600. The use of the selectable crystal oscillator is seen on some "customized" SP-200 WWII Super Pro receivers and was even designated as an "Improvement Kit" with the part number MC-531 from the Signal Corps. Sometimes these receivers were "custom" rebuilds done by various engineering companies for the Signal Corps. These "custom" SP-200 Super Pro receivers usually date from 1947 up into the 1950s. The turret band changing was also a specification from the Signal Corps and this is shown by the Hallicrafters' R-274 receiver, built for the Signal Corps and sharing many similarities to the SP-600, including the turret band switching concept. So, although Hammarlund advertising appears to ignore the Signal Corps input, there is just too much evidence available that shows much of Hammarlund's design concepts for the SP-600 actually came from the Signal Corps. The first contract issued was in 1950 for the R-483/FRR for the Signal Corps and was actually a SP-600 JX-5 and was not actually built until November 1951. The first built SP-600s were R-274A/FRR receivers (SP-600 JX-1) built on a 1951 contract in September 1951.

Most SP-600 versions were built throughout the 1950s but the SP-600 continued to be produced in smaller numbers up into the early 1970s. The standard SP-600 tunes from .54 to 54MC in six bands. A "J" suffix indicates JAN parts were used in the construction and an "X" suffix indicates a switchable crystal LO. Hammarlund also offered a "JLX" version with 100-400KC substituted for the .54-1.35MC band and a "VLF" version that covered 10KC-540KC. Hammarlund made over 40 variations of the SP-600 that were each assigned a numerical suffix which identified the particular circuit, mechanical changes or sometimes the end-user. The last in the "time-line" was the model variation SP-600 JX-21A from 1969-1972 which utilized a product detector circuit, two additional tubes and some other changes to make it "compatible" with SSB operations.

Most versions use a 20 tube double conversion superheterodyne circuit with a rotating turret bandswitch. The receivers also feature enormous proportions, robust construction and oversize controls - along with a super-smooth tuning system that only adds to the enjoyment of operating these fine receivers. Double conversion is switched in above 7.4MC and uses a 3.5MC crystal-controlled conversion oscillator. Though the SP-600 has two dials, it has no bandspread - the right side dial is a logging scale allowing precise retuning of desired stations. On the left is the main tuning dial and the mechanically articulated dial pointer that indicates which tuning scale is in use (along with the small window between the dials that shows which tuning range is selected.) The tuning arrangement was an up-dated version of the "Continuous Bandspread" system introduced in RCA's AR-88 series receivers in the 1940s. The frequency readout accuracy is vague which is why a precise logging scale system is incorporated into the SP-600 design. The Carrier/Audio Level meter is not illuminated and a switch is provided to indicate either carrier level or audio output. Most (but not all) SP-600 model numbers usually will have a suffix with "J" or "JX" followed by a numeral. As mentioned above, suffix "J" indicated that, as much as possible, military level components and construction were used. Suffix "X" indicated that a switchable six-position, fixed-frequency crystal-controlled oscillator was installed that allowed the user to install HC-6/U type crystals for specific desired LO frequencies. The user could switch to any of the crystal LO frequencies for increased stability for that particular frequency - however, the receiver still has to be "tuned" to the desired frequency for the RF and Mixer stages to be in tune. The number following the suffix generally indicates specific features for that version, e.g., contract or end user, circuit changes, etc., with the number ranges being more or less chronological for introduction until the last of production. Though the numbers assigned are chronological for their introduction, many of the numbered versions were produced over a relatively long period of time that saw many engineering changes incorporated into the receivers. As an example, early 1953 JX-21 receivers are different from the 1956 USAF documented JX-21 and both of these versions are very different from the JX-21A of the late sixties. Many SP-600 receivers were set-up for diversity operation and the standard diversity model was the JX-17 version. This version was produced in large numbers and can be easily spotted by observing that it has two extra controls and uses three red colored knobs. The SP-600 Audio output is about 2 watts from a single 6V6 using a balanced split-winding audio output transformer for 600 ohms Z. A rebuilt SP-600 will provide an impressive "communications grade" type of audio reproduction.

Shown in the photos left and right are two advertisements for the SP-600. On the left is the initial 1948 ad for the Series 600 "Super-Pro," the SPC-600X. This was a "teaser" ad to show what Hammarlund was working on. It's possible that a prototype or two may have been made but the SPC-600X was never released as a sellable product. To the right is an ad for the SP-600 from 1963. By this time, Hammarlund was owned by Giannini Scientific Company and the products were built at Mars Hill, North Carolina. Note that the 1963 ad promotes the SP-600's wide frequency range (although you would have to buy two receivers to have the advertised coverage.) Also note that by 1963, it appears that only four SP-600 versions are being produced. The influence of the earlier Super Pro receivers on the SP-600 design and construction is obvious. In 1931, Hammarlund sky-rocketed to the top of communications receiver technology with the introduction of their Comet receiver, the first successful shortwave superheterodyne receiver. This was followed by the Comet Pro receiver. Hammarlund jumped to the top again with the introduction of the Super Pro in 1936. The Super Pro remained basically unchanged from 1939 up to the late forties and, by that time, it was apparent to most users that the SP-400 Super Pro was just a repackaged SP-200. The Signal Corps wasn't interested in the SP-400 Super-Pro and preferred to modify WWII BC-794 (SP-200SX) receivers rather than buy the new Super-Pro. The Signal Corps was waiting for Hammarlund's new receiver - one that they had significant influence over its design - the SP-600 receiver. Once again, Hammarlund had to come up with a tremendously innovative receiver and, in 1950, they once again "hit it out of the park" with the SP-600. The new receiver became a favorite of both military and commercial users. Nowadays, many rebuilt SP-600 receivers have found new homes in vintage AM ham stations around the world. Admired for its massive proportions, ultra-smooth tuning, great sensitivity and sheer dominance of the ham station landscape, SP-600 receivers are sought-after, rebuilt, restored and then finally, once again, put to work.

The Molded Capacitor Problem

photo left: Top view of the 1953 SP-600 JX-21 chassis.

Some Preliminary Rebuilding Information

The Caveat - Though most SP-600 receivers look pretty much the same on the exterior, inside there are several differences between the many variations with different value components, additional parts or lack of parts being commonly encountered. The receivers shown in this article are two different versions of the early variety. The SP-600 JX-21 has the switchable crystal controlled oscillator and all of the modifications to incorporate the "JX" option. This is an early version (1953) JX-21, so it is somewhat different than the later 1956 USAF versions and completely different from the 1972 JX-21A version. The other receiver is the SP-600-25C. This receiver has the 25-60 cycle power transformer and lacks the switchable crystal controlled oscillator. There are several minor differences between these two types of SP-600 receivers and certainly many differences with the other 38 or so variations of the SP-600. So, if your SP-600 doesn't exactly conform to what is stated here, don't be too surprised - I won't be either.

 Manuals and Schematics - There are about 40 different versions of the SP-600 but the receivers can be roughly divided between early versions and later versions. This division is not based on the number suffix but rather when the particular receiver was manufactured. There aren't manuals for every single numbered version but it's fairly easy to find a manual and schematic that is proper for your particular version.

The Military came out with several different manuals for many of the variations, so it's possible to find both Signal Corps TMs (TM11-851) and several types of USAF manuals. However, with some of the military manuals you will find there are earlier manufactured versions of the receiver than what is covered in the manual. This is particularly true of the JX-21 USAF manual which is dated 1956 and is for the later versions of the JX-21. The JX-21 was first built in early 1953, which means the 1956 USAF manual will have several differences from the early JX-21 receivers. This is true for almost all SP-600 receivers versus the available manuals. It is more important to use the manual that was issued closest to the date of manufacture of the receiver rather than going by the particular numerical version of the receiver. In other words - not all specifically numbered versions of the SP-600 were identically built.

With the early versions of the schematic as in Issue 1 and 2 of the Hammarlund manuals no component values are on the schematic. You'll have to use the parts list as a cross-reference to determine the correct value of the component shown on the schematic. Later schematics, from Issue 3 up, do have the component values on the schematic.

BAMA (Boat Anchor Manual Archive) has about four of the six issues of the Hammarlund manual online plus other SP-600 information. Generally, Hammarlund Issue 1 and 2 are for the earliest versions while Issue 3 is sort of a transition version. Issues  4, 5 and 6 are for the later versions. This is generally the case and it's best to down load all of the versions available to see which is closest to your particular SP-600.

There are some dealers doing business on eBay that sell repro manuals if you are interested in having hard copies of the Air Force issues and of the Signal Corps TM (around 200 pages, so it's expensive.) If your SP-600 was modified or has missing components you will absolutely need to use the manual and schematic that is correct for your receiver. Though it is possible to work through the SP-600 with general information from almost any manual, if the receiver has been compromised with modifications, you'll need accurate documentation to return the receiver to its original configuration.

photo above:  How the typical SP-600-JX receiver looks before restoration. Lots of dirt and corrosion on this one.

Photo above: A look under the chassis shows a lot of molded capacitors (even some replacement caps that are molded types.) Other serious problems are surely lurking inside this typical unrestored  SP-600 JX.

1. RF Platform - 20 molded capacitors - all are .01uf - 400wvdc

2. Turret Bandswitch - 6 molded capacitors - all are .01uf - 400wvdc - only the 1RF, 2RF and Mixer modules for the .54-1.35mc band and the 1.35-3.45mc band have molded capacitors that need to be replaced.

3. T-1 455kc/3.559mc Mixer Transformer - 1 molded capacitor. This unit is mounted to the side of the RF box next to the filter chokes. Inside the unit is one .01uf - 400wvdc molded capacitor.

4. Crystal Conversion Oscillator - 3 molded capacitors. This unit is mounted under the chassis. All capacitors inside are .022uf - 400wvdc.

5. IF Transformers - 2 molded capacitors inside the Conversion IF transformer, value is .022uf - 400wvdc. 1 molded capacitor inside two of the 455kc IF transformers, value is .022uf - 400wvdc. The Crystal Filter/IF transformer doesn't have molded capacitors inside.

6. Switchable Crystal Controlled Local Oscillator Assembly ("X" Option) - 2 molded capacitors inside this unit. Both are .01uf  - 400wvdc.

Additionally, most receivers will have two tub units that are double .05uf at 600wvdc capacitors. There is also a .05uf  tubular and a .25uf tubular (Sprague built units usually) that are mounted on standoffs on the front apron of the chassis. The usual procedure is to replace these with individual components when doing the rebuild. All of the other under-the-chassis molded capacitors are easy to access and replace. All total, the amount of molded capacitors that need to be replaced will exceed 50.

Components - When Hammarlund eliminated the molded capacitors from production they went to ceramic disk capacitors. That choice as new replacements would be fine, as would SBE Orange Drop or metalized film "Yellow-Jacket" types of replacement capacitors. The choice is not critical since any of these types of capacitors are at least a 1000 times better than the original molded capacitors.

Rebuilding the SP-600 Assemblies

The RF Platform

This is the most difficult assembly to remove, rebuild and then re-install. The bandswitch must be placed in a position exactly in-between any of the "lock-in" band select positions. This allows the short contact pins from the RF units in the turret to not be engaged into the contact fingers of the RF Platform. Remove the top cover over the tuning condenser and unsolder ALL of the connections to the tuning condenser. There are 16 connections that need to be de-soldered here. Then remove the top cover from T-1 and unsolder the wires that are connected to the terminals there. There are from five to seven wires that need to be unsoldered depending on which version is being worked on. Next remove the four mounting screws on the top of the RF Platform. Now, carefully lift the RF Platform up while guiding the T-1 wire cable as they pass through the opening to T-1. With the RF Platform out of the receiver, you'll be able to see 20 molded caps that need to be replaced. Be sure to watch the alignment holes versus the placement of the new capacitors. You want to have clearance from the alignment holes down through the holes in the ceramic mounts and down to the adjustments on the RF units of the turret. Make sure your placement of the new capacitors doesn't block alignment tool access. Try to mount the new capacitors in exactly the same position and same orientation as the originals. Check the contact fingers for corrosion and clean with De-Oxit applied with a Q-tip - nothing abrasive. Don't move the bandswitch control on the receiver until you have reinstalled the RF Platform. Re-installation requires that the wire cable be guided through the access hole to T-1 as you lower the RF Platform into place. Reconnect the wires to the T-1 terminal board and reconnect all 16 connections to the tuning condenser. Replace the mounting screws. Check for proper operation of the bandswitch-turret and smooth engagement of the contact pins of the turret into the RF Platform contacts.

photo above: The RF Platform showing the original Cornell-Dubilier molded capacitors. (SP-600-25C)

photo above:  Close-up of the broken contact finger on the RF Platform, 1st RF stage, Antenna input. These contact fingers are very fragile at the points where they enter into the ceramic mounts. The right-angle bends tend to be weak points.  Hammarlund soldered the bottom lugs on the contact fingers to lock them in place and to prevent "rocking" as the turret pins are engaged.  (SP-600 JX-21)

Removing the contact finger requires careful workmanship. You must use solder wick and remove absolutely all of the solder from the contact finger lugs. Next, insert an 18ga. wire into the "T" slot in the ceramic and push gently down on the tang that holds the contact finger in the ceramic. Once the tang is down, that side of the contact finger can be pushed slightly up. Insert the 18ga. wire in the other "T" slot and push down on that tang, then that side can be pushed slightly up. Once the two tangs are released the contact finger can be removed from the top of the ceramic mount. These contact fingers are very fragile and will break easily, so go slowly and carefully, then successful extraction is fairly easy. To install the good contact finger just push it into place in the slots and the tangs will lock. Then solder the bottom lugs to lock in place.

One can also note from the close-up photo to the left that there is some corrosion on the fingers and also a lot of dirt or "tuner grease." This should all be cleaned with denatured alcohol and a Q-tip. Stubborn areas may require a brass brush to remove the corrosion. When everything is clean then go over the contact fingers with De-Oxit D-5, which contains a corrosion cleaner along with a lubricant that works fine for these kind of contacts.

Careful, meticulous rework to the RF Platform is essential to regain the superior performance that the SP-600 is capable of.

Another interesting observation is that R-16 is missing from the circuit. This 1K ohm resistor is the plate load for the 2RF amplifier. Looking at the photograph to the right, R-16 should be connected to the terminal to the rear that R-15 is also connected to. It can be seen that no resistor is mounted there. During the initial check out of the receiver, operation appeared normal (for an original condition receiver.) So one can conclude that even though the 2RF amplifier was not operating to its full capabilities, it did pass enough of the signal through that stage that the defect was not noticed in the initial operation check out. This shows that a thorough physical inspection of the RF Platform is necessary to find all of the defects that might be present. The Connection Diagram for the "Tuning Unit" (RF Platform) is extremely helpful for the inspection.  

photo right: Close-up of the Mixer section of the RF Platform which is where the plate circuitry is for the RF2 amplifier. Note the discoloration of R-15 and the missing R-16. (SP-600-25C)

photo above:  The band switching turret as viewed from the underside of the chassis with the shield-cover removed.  Note the curved retaining clips on each end of the RF modules.  (SP-600 JX-21)

The Bandswitching Turret

This is a really great piece of engineering. As the bandswitch control on the front panel is turned, the turret rotates and places into an engagement position the contact pins of four front-end modules that make up the 1RF, 2RF, Mixer and LO sections of the front-end for each tuning range of the receiver. Each front-end module is built onto a ceramic mount that has the tunable coil, the trimmer capacitor and a couple of resistors and fixed capacitors that make up each circuit. The connections are to short pins on the back of the ceramic mount that engage the RF Platform fingers when the turret is rotated into position. Each front-end module can easily be removed from the turret since they are only held in their mounted position by two arched clips.
 

Six of the RF modules use molded capacitors that will need to be replaced. All of the others have mica caps installed. The .54mc-1.35mc section will require replacement of a capacitor in the 1RF, 2RF and Mixer modules and the same units require a capacitor replacement in the 1.35mc-3.45mc band. Each module has one molded capacitor of .01uf at 400wvdc. While you are working on the turret, it's a good idea to check the resistors along with the other components installed in all of the front-end modules. The Mixer sections especially seem to sometimes have burned resistors but they do turn up in the other modules also. Once and a while you'll find someone has inadvertently injected high level RF into the receiver antenna input and burned the 1RF coil. Check the pins of each RF units for corrosion and cleanliness. Don't use anything abrasive unless corrosion is present, then you might have to use a brass brush and denatured alcohol to clean. Follow with De-Oxit applied with a Q-tip.

Be sure to check the tightness on all of the tensioners that are on the inductance adjustments. These are circular metal disks with slots along the edge that are threaded onto the inductance threaded shaft. These tensioners provide a spring load on the inductance adjustment so it will stay set. Sometimes these tensioners loosen and when that happens you'll find that the inductance adjustments don't seem to hold adjustment when doing your alignment. Check them - they should be a little more than snug but not over-tight. Too tight and you'll have difficulty adjusting the inductance with the alignment tool. The tensioner should be just tight enough to hold its position and not move when the inductance shaft is adjusted.

NOTE: Early modules will have different value resistors compared to later ones.

Photo left:  RF front-end module with the original C-D molded capacitor installed (SP-600-25C)
 

photo above: 2RF module (3.45-7.4mc) with burned 510 ohm resistor (SP-600 JX-21)

The IF Transformers

There are molded capacitors located in each of the IF transformers with the exception of T-3, the crystal filter/IF transformer assembly.  IF transformers T4 and T5 have one .022uf 400wvdc molded capacitor inside that will need to be replaced. T-2 has two .022uf 400wvdc molded capacitors inside that will need to be replaced. The IF transformer shields are held in place by four 6-32 screws that mount the shield to the chassis by way of right angle brackets on the shields. Removing the screws and the shield allows you access inside of the IF transformer. There is a blade spring that provides a load on the IF adjustment screws. Be sure to keep track of the spring and replace it when you re-install the shield. The molded capacitors are very easy to replace in all of the IF transformers. Be sure to check the value of the carbon resistors that are located inside the IF transformer shields since any leaky capacitors may have compromised the resistor value.

NOTE: Be aware that the IF transformers change slightly in the later versions of the SP-600 and will have slightly different connections to the terminals under the chassis.

photo right: The IF section of the SP-600 with the shields removed allowing access to the molded capacitors. The nearest IF transformer has already had the new SBE "Orange Drop" replacement capacitor installed. Note here that the 2200 ohm resistor is 5% - I had to replace the burned up original 10% resistor because of the leaky original capacitor (SP-600-25C.)

photo above: T-1 455kc/3.955mc Mixer Transformer with the shield removed showing the C-D molded capacitor located inside. This unit requires the removal of a couple of other adjacent assemblies just to gain access to replace the one molded capacitor located inside.  (SP-600-25C.)

T-1 455kc and 3.955mc Mixer Transformer

This is the small assembly that is mounted on the right side of the RF tuning box near the power supply filter chokes. It is held to the RF box by six screws that are difficult to access in the  non-"X" receivers because of their proximity to the power supply filter chokes and impossible to access in the receivers with the "X" option. In "X" optioned receivers, the Crystal Controlled Oscillator unit has to be removed first. Unsolder the four wires that connect to the Filter Capacitor/Filter Choke mounting assembly, remove the four screws and dismount the assembly. You don't have to unsolder the wires that connect to the Crystal Controlled Oscillator unless you are going to remove it for rebuilt, then unsolder them. Unscrew the fours screws that mount the Crystal Controlled Oscillator and move it out of the way (or remove it for rebuild if you've unsoldered the wires.) Now you should have easy access to T-1. On non-"X" receivers all you have to do is remove the mounting screws from the nearest filter choke and move it back and to the side to allow the cover of T-1 to be removed. Once the cover of T-1 can be removed, then you can access the single .01uf 400wvdc molded capacitor located inside. After replacement of the capacitor, the cover can be re-installed. If you are not going to rebuild the Crystal Controlled Oscillator then remount it (but if you have then resolder those wires) and then remount the filter choke-capacitor assembly and resolder those wires.

The 3.955mc Conversion Oscillator

This assembly is located under the chassis. The shield is mounted to the chassis using four 6-32 screws. Removing the screws and the shield allows access to the 3.955mc oscillator circuit. There is a vertically mounted fiber board that has some components mounted to it. There are three .022uf 400wvdc molded capacitors used in this circuit that will need to be replaced. When re-installing the shield note that there are three notches that allow the circuit wiring to pass under the shield when mounted without pinching. Be sure the wires pass through the notches. Also note that there is a notched fiber spacer that sets on top of the vertical fiber board. This must be in place when re-installing the shield since it prevents vibration movement of the vertical board that might mechanically affect the 3.955mc crystal.

The Selectable Crystal Oscillator ("X" option)

This unit allows the operator of an "X" optioned SP-600 receiver to substitute a crystal LO for the regular LO for increased stability. Six crystal sockets allowed the user to install HC-6 crystals for fixed frequency operation. A six position switch allowed selecting any of the crystals installed. There is also an ON-OFF function (labeled "VFO") that routes the B+ from the standard LO to the JX oscillator. A vernier frequency control was also provided. The VFO/crystal channel switch and the frequency vernier were brought out as two extra controls added to the upper right section of the front panel.

When selecting a Crystal Controlled LO frequency, only the LO is changed - you will still manually have to tune the receiver to the selected operating frequency in order to have the RF and Mixer stages also tuned correctly. You'll need to install the correct frequency HC-6 crystal which is not the same frequency that you want to tune in. The manuals will provide the formula to calculate the crystal frequency required for the desired receive frequency. Basically, 455kc or 3.955mc is added to the desired receive frequency to obtain the correct crystal frequency depending on the band and whether the receiver is operating single or double conversion. Also, on the higher frequency bands the result is divided by 3 or 4 for operation on the crystal frequency harmonic.

Removing the Crystal Controlled Switchable LO unit requires removing the knobs and any shaft couplers. Remove the filter choke/filter capacitor sub-chassis to access the wiring harness connections. Five wires have to be unsoldered there. There is also a connection to the switch and the RF Platform to unsolder. Remove the screws from under the chassis that mount the hex stand-offs that are attached to the bottom of the unit.  You should now be able to extract the entire unit from the receiver. Several screws need to be removed (and the sheet metal cover sides slightly bent outwards to remove it) in order to gain access to the Crystal Oscillator. Once inside you'll find two molded capacitors that need to be replaced. Check the resistors for drift and clean the switch to complete the rebuild. Re-assemble and re-install in the reverse order.

Once the Crystal Oscillator unit back in the receiver and is operational, it's doubtful this particular option will ever be used unless you want to listen on one frequency - exactly. When the SP-600 is running on its own LO it doesn't drift after a 10-15 minute warm-up, so only the most demanding frequency-stable military data-type (RTTY usually) requirements would benefit from using the Crystal Controlled LO.

photo above:  Inside the Crystal Controlled LO unit. Note the two orange-drop replacement capacitors behind the RF chokes. (SP-600 JX-21)