AN/GRC-9 Radio Set

Updated 8/1/2024:

US Marines of a Tactical Air Control Party with the First Marine Division in combat at the Chosin Reservoir operating an AN/GRC-9 transmitter-receiver.  Hand cranked generator, cold weather, possibly watching Marine Corps Corsairs or Panthers rolling in on target:  NK or ChiCom troops danger-close, Nov-Dec 1950.  Hopefully it got REAL HOT for the Reds in short order. Field day of the highest order, not fun.  But I bet it did the job…

Note the weapon (.50 Cal BMG) cleaning rod likely being used for a ground rod. They are using the older GN-58 hand generator; the receiver battery cord connector is visible. Note the plywood pack board below the “crankers” left leg, probably used to hump the radio or generator. The transmitter band switch appears to be set to Band 2 Master Oscillator (3.6 to 6.6 mc range.)

The Marine holding the T-17 mic and map is likely a pilot from the supporting Marine fighter squadron. The Marines found that this type of assignment produced the best air-ground, close air support coordination possible.   Thanks guys. We remember.  So do the South Koreans of the Republic of Korea.
Official USMC Photograph.

The GRC-9 was widely used during the Korean War.  Below: Anti-communist North Korean partisans operating a GRC-9 and hand generator.  They were highly motivated volunteers from the tens of thousands of Koreans who fled south after the communists seized control of the north.  They were trained, equipped and advised by US Army Special Forces personnel to help liberate their homeland.

Self-named as “Donkey” units, they conducted raids and unconventional guerilla operations, taking the war back to the Chicom and North Korean communists in coastal North Korea.

A possible origin for their name was put forward by an early Donkey leader who stated “the generator of the [AN/GRC-9] radio looked like a Korean donkey or ass. When you crank the generator… you have to ride on the generator which looks like a rider on the back of a donkey.” (Reference 105).

Or possibly a simpler origin in the Korean word Dong-il (Leader). There were approximately 12,000 of them organized into 15 Donkey/Wolfpack units.

The GRC-9 in AM Voice mode would be quite effective for communicating with their Team Leopard base station on Paengnyŏng-do island off the west coast of North Korea.  They had been trained in the use of One Time Pads.

“With voice transmissions ruled out by distances, communications specialists must rely on continuous wave radio operation, using the International Morse Code.  They must be able to receive and transmit at the rate of 18 words per minute, perform maintenance up to Third Echelon and even build a working set out of parts at hand.  Other team members must be able to work at least five words per minute and know how to set up and operate a radio.  Workhorse of the Special Forces is still the dependable Angry Nine, officially the AN/GRC-9.”
Special Warfare U.S. Army, Chief of Information, U.S. Army 1962.

Later on, same problem, same idea, same radio, different place:

Above:  The “GRC-9” at work with U.S. Army Special Forces and Vietnamese irregular forces fighting the North Vietnamese Army and Viet Cong communists in South Vietnam.  Note the taller G-43 generator had replaced the GN-58 in this instance. Official Photograph, U.S. Army, August 8, 1964.

That image is from a Signal Corps film following a Special Forces patrol in Vietnam. The filmer or the person that entered the description in the National Archives file mis-identified the radio as the GRC-109. Incidentally, this 1964 photo is the only one I have found of US personnel using a GRC-9 in Vietnam. Well into obsolescence by then, the GRC-9’s were used by some US advisor personnel, especially early in our involvement.

Shown here, likely training and supporting an irregular forces patrol, probably Montangnard villagers . However, the GRC-9 was widely used by the South Vietnamese government Civil Guard and Self-Defense Corps in the early 1960’s and also by ARVN units as well. The Viet Cong captured many of these radios and used them themselves.

The ChiCom Type 102E set is a repackaged, Chinese copy of the GRC-9 with a few changes. The North Vietnamese Army and VC used the Type 102E extensively in their assault on the south.

WWII? The AN/GRC-9 was a mid-late 1940’s development as an incremental improvement and replacement for the SCR-694/BC-1306 which itself was a replacement for the SCR-284/BC-654. Due to a lack of any actual deployment evidence to the contrary, it is widely understood that the GRC-9 did not see service in WWII as the war ended (VJ Day) before it was available or even ready for issue.

Here is an interesting shot of an SCR-694 being used by OSS Special Operations personnel on a rice paddy dike in China in 1945.  It’s not known if it was before or after VJ day 1945.

That would indicate kind of a big change of direction on that date regarding our support of Mao & Co. against the Imperial Japanese Army or Chiang Kai-shek against the IJA AND Mao’s Communists shortly thereafter. (Reference 105)

A listing of war supply contracts (Reference 101) shows that a $90,000 contract, SC 921, was awarded to Crosley Corporation for GRC-9 radios in June 1944.  It further shows that the “Contract Completion” was in January 1945. The report does not indicate what Contract Completion actually means. (Pertinent signatures obtained?)

However contract dates and subsequent procurement orders, production, receipt and fielding can be quite different. As WWII drew down many late war contracts were cancelled but we don’t know much about this contract or any associated, possible production run.

Other detailed US Army WWII procurement records were published by the Office of The Chief of Military History on 9 April 1952 (Reference 74).  It shows that during 1940-1945 no procurement’s of the AN/GRC-9 were made during that time frame.  To me, further evidence that they did not see service in WWII.  The SCR-694 was being procured in 1944 and 1945 for 14,416 and 9,316 sets respectively to replace the SCR-284’s.

Orders were made in 1951 and 1952 to 4 different manufacturers of the GRC-9. These included Lewt, Crosley division of AVCO, Tele King and Hoffman Radio Corporation. Per unit prices varied from approximately $1,256 to $1,605 depending upon the year, manufacturer and contracted quantity. (Reference 78).

A total of 63,972 SCR-284 sets were procured during its run but none in 1945 as the SCR-694 took over and the end of the war was in sight.

Below is a photo of a 10th Special Forces Group LT showing their gear to a MGEN and some 5th Army and 82and Airborne officers in Bad Tölz Germany.  Photo Undated. The LT has what might be the “Trojan Horse” insignia on his beret upon close examination.  That insignia dates the photo as being taken sometime between December 1955 and 1962 when it was replaced.  Army uniform trivia used to date photos!)

Notice GRC, C’s and Ski’s.  (Plus an M-3 SMG and M-1/M2 Carbines). In this case the GRC is a GRC-9, left foreground.  Present is the GN-58 hand generator along with the generator, antenna and accessories carrying bags BG-172, 174 and 175.

Incidentally, SF units like this could have also used the CIA’s RS-6 or RS-1 HF/CW sets in the 1950’s, then also available to US military.  The RS-1 then being replaced by the Army’s nearly identical GRC-109 in the mid-1960’s.  Ideal for these types of missions, they were better suited for man-portable Ops than the GRC-9 for many reasons.

The radio in the above photo is mis-identified in a subsequent Army Special Operations Forces historical publication as a GRC-109 just like the same mis-identification of the GRC-9 in the Vietnam photo above. US Army Photo.  (Reference 105).

The GRC-9 did see widespread service during the Korean war and well into the 1960’s by US forces. One interesting application was in AN/GSS-1 Electronic Search Central mobile shelters as part of HAWK SAM missile units to coordinate missile battery operations.  Those Command and Control shelters also included the AN/GRR-5 HF Receivers.

The GRC-9 was also used in NIKE SAM missile batteries for site-to-site communications.

It was also widely used by our European NATO allies, probably well into the 1970’s, notably by the French, German, Dutch and others. Germany and France also built copies and made system improvements for their applications, notably the development of external power amplifiers. GRC-9’s also saw considerable service in Africa, Indonesia and many other areas around the world.

Its principal tactical use was in relatively short range networks that were beyond the range of the emerging FM voice sets coming into play at the time. With a planning range for CW of 30 miles ground/20 miles vehicular and 10 miles voice (TM 11-263 et al) it could meet that need. It could of course work over much further distances as the situation and conditions warranted but that was probably not its intended role.

Above: US Marines with Second ANGLICO equipped with the GRC-9, GN-58 and a PRC-8/9/10. Almeria, Spain. Circa 1958. This Op possibly related to the US support to the government of Lebanon combating a 1958 Muslim insurrection. Photo: SGT Ed Nunn USMC.

So how long did the GRC-9 remain in US military service, or in the hands of allies or Reserve units for that matter? Not much information out there. Take a look at this shipping box stencil:

This set was held at the Sacramento Army Depot (California) in the early 1970’s and was still “in the system”, tested and presumably ready for issue in April 1974. Note the “00” Country Code in the National Stock Number. This is the 2-digit code for United States (origin) and was added to the former 11 digit Federal Stock Number after 30 September 1974 as part of NATO supply and logistics standardization.

The “00” implication is that dropping the GRC-9 from US DOD stocks was a decision that came some time after 30 September 1974.

(NSN 5820-00-543-1996 identified the VRC-34, the vehicular mobile configuration which includes the RT-77 receiver-transmitter as the main component. The VRC-34 also includes the dynamotor power supply and all necessary accessories for vehicular operation.)

In Civilian Clothes:

There are a lot of GRC-9’s in the hands of European, South American and US Hams where they still enjoy a enthusiastic following. They are versatile, rugged and dependable and are readily put into Ham service without the dreaded “conversion” or “Ham-mering”.

Below is my “GRC-9” set up on the Pacific beach at Spooner Cove near San Luis Obispo CA during less stressful times.  (Actually, it is a GRC-87 – do you know the difference?*) We were running “Coastwatcher Ops” and transmitting Bikini Reports back to Net Control at Camp San Luis Obispo.  This was during the Military Radio Collectors Group meeting in May 2010.

(*The RT-77/GRC-9 receiver-transmitter when deployed with the G-43 generator, LS-7A (or LS-203) speaker and the ME-61/GRC field strength meter is defined as the AN/GRC-87 “system”. (TM11-5820-453-10, 10MAY1963 refers) )  If you’re keeping score.

The distance from here to Camp SLO NECOS was 9.2 Miles over hilly, rocky terrain; they were running a TCS set.  From here to another station in this 3-station Net was 3.3  miles, and from there to Camp SLO was 5.8 miles.   That station was running another GRC-9 in a portable configuration however he was using the DY-88 Dynamotor power supply.

Under the conditions at that time (appx 1000 hours local) the signal to noise ratio of received signals was 2 to 3 times better on the whip than the AT-101 wire which surprised us.  It was running approximately northwest, broadside to Camp SLO, averaging about 6 feet off the sand.  CW and Voice Comms with the 15′ whip antenna were Q5.

I didn’t note the FoF2 critical freq at that time to see if the low wire would work in NVIS mode under those conditions.  Apparently it was lower than our 3885 freq and/or the D Layer absorbtion was very high then. So little/no skywave contribution.

We ran the standard ground radial wire array, G-43 hand cranked generator and a battery for the receiver to save the “power supply”.  The TCS was running a 30 foot inverted L and the other GRC-9 was running the standard whip antenna.  This setup would have been a very workable tactical circuit. Authors Photo:

Below is a typical mountain top campsite to cover all circuits. There is the GRC-109, GRC-9, PRC-47 and the mobile GRC-9 all on-line. However, looks like the flash caught the Midwatch off guard. Zzzzzzzz.  Good thing the Sergeant wasn’t around!

Below is a shot of our Perimeter Defense Force checking out the Company Command Post.  Note the GRC-9 AC Power Supply built in the .50 Cal Ammo Box on the ground.  Provides all voltages from 120 VAC.

Above: The GRC-9 was operational at the Military Vehicle Collectors of California April 2013 Rally at Camp Delta. A Battalion Communications Center capable of handling comms relay between any possible Time Warp from 1944 through the 1970’s. We worked lots of stations on 40 meters CW with the GRC-9 and its’ simple whip antenna. Always a hit with the visitors.

Above: Later that night…..

Don’t let this happen to you! Do your PMS and then double-check everything!

Above: QSL Card from the USS Pampanito SS-383, normally moored in San Francisco near the Golden Gate bridge. I worked their ham station, NJ6VT aboard the sub on CW with my GRC-9 from our Forward Operating Base in the Sierra Nevada mountains. Her wartime Navy call sign was NJVT.

Below is a shot of the GRC-9 in the field using the .50 caliber AC power supply. “You have a strange rushing sound in your Transmit audio”….Yup…. This is the setup I used to work the USS Pampanito submarine on CW during a “Museum Ships On The Air” weekend. Better than 150 miles on 40 meters during the day.

Below is another shot while sending Ditties from FOB (Forward Operating Base) Margarita….The little AC power supply in the .50 Cal ammo can makes it convenient to work in camp. Especially when you bring the generator to run the Blender anyway!

Below: Would you believe a solar-powered GRC-9? Been there, done that. Here we are set up in the mountains west of Lake Tahoe at OP COMMANDO, monitoring kayak traffic 2000 feet below us on the North Fork of the American River. Panel charged the deep cycle battery which powered the sine wave inverter that powered the ammo box power supply which powered the GRC-9. Inefficient as hell but it worked!

Below is another shot of that same position – we also had a PRC-47  going on CW and LSB as well. There is the wreckage of a C-46 Commando somewhere within a mile of this location.  Very dense forest in here.  It’s on our list for a search expedition. This spot will make for a very good base camp.

Above: Mounting my other GRC-9 in the back of the old Bronco makes it pretty convenient while camping. I can check into the Saturday night West Coast Military Radio Collectors Group net on 3985 KC AM, even from my sleeping bag when it gets cold up here in the mountains. I can even get the latest soccer scores from Radio Australia up here.

Many “gray hairs” later, it’s still going strong at campsites.  I’ve put a lot of miles on this old set; never fails.  It has all matching serial numbers, 2974.

I usually run 80/40 meter coax-fed “fan” dipoles but sometimes just a quarter wavelength wire or the MP-57/MS vehicle-mounted whip depending upon how long I plan to camp in one place. If the raccoons wake me up at 0345 I can check the propagation worldwide without getting out of my Extreme Cold Weather sleeping bag.

The Bronco is pretty cramped with camping gear so I had little room for a huge DY-88 dynamotor power supply. So I built a small transistorized power supply that runs off the vehicle 12 Volt battery; with the little supply sitting under the passenger seat. It works well and is more efficient than the dyno but I miss the Dyno Whine during Transmit !

Above: The GRC-9 at work down in the Commo Bunker sending out the nightly SITREP at Zero Dark Thirty.
(BTW, it’s an old military phrase that has been around since clocks were invented and the troops had to get up early. Then the Hollyweird crowd discovered it and was so enthralled with it they named a movie after it…. Same with “Hurt Locker” but I digress…Those people need to get out more often – yeah, right.)

UPDATE: 60 meter trials with the GRC-9 on CW:

I set up the GRC-9 on the 60 meter Channel 4 CW frequency of 5373 kc. Driving a 40 foot random wire in the trees from the Commo Bunker.

Experiments on 60 meters CW: It worked well, getting into the KPH SDR receiver about 50 miles from here, mid-day, at an S-7 or so signal strength. Keying dynamics and drift stability were both acceptable on that frequency. I used a counter to set the frequency to mid-channel as required.

More experiments, remote contacts and field ops are in order.
—————————————————————————————-
A note on the GRC-9 home brew power supplies discussed above: I have gotten many requests for the design of the various AC and DC power supplies that I have built. (These are “one-of” power supplies for my own use – I don’t manufacture or sell them.)

As we know, the GRC-9 requires a number of different voltages from the power supply. 580, 108 Reg, 6.3, 1.5 ……I was able to build supplies around junk box parts I had on hand, but the design is entirely dependent upon the kind of transformers you can get – it is central to any design you could make. It completely defines the circuit specifics you will need.

The AC transformer I used in the Ammo Box supply is from an old oscilloscope PS,  that “TV” transformer is a Triad R-50BC – long obsolete. It had lots of different voltage windings. Along with a big choke it was a challenge to package everything into that small box.

The DC supply is built around a transformer from an old Heathkit mobile transceiver DC supply that was scrapped – and it also had a lot of winding voltages I could work with and adapt. Especially the all-important primary and feedback windings needed for a 12 VDC transistor power oscillator.

Above: The GRC-9 set up on the workbench along with the “Ammo Box” AC power Supply. A small, simple, compact solution to powering the GRC-9 from AC sources at home or in the bush via a generator.

Without either of the AC or DC oscillator transformers my design would not be useful to anyone trying to replicate them. That said, however, the designs are otherwise straight forward and follow the ideas in any of the older ARRL handbooks for providing those voltages. I used adjustable LM-350 voltage regulators for the +108 and 1.5 volt supplies.

I chose to regulate the receiver filament voltage to protect the tubes and I also regulate the receiver B+ in the supply. Consequently, the VR105 (OC3) tube in the GRC-9 does not “fire” with externally regulated +108V applied and this works well.

Unfortunately no one makes suitable (affordable) HV power supply transformers any more in the “transistor age”. Especially transformers for DC-DC HV converters. You must work with what you can find – and that took some digging….
And that was part of the fun……

Above: The GRC-109 DC-DC Power supply built into an excess GRC-109 transmitter box. This supply powers the GRC-109 transmitter and receiver from 12 VDC. Since both the GRC-109 receiver and the GRC-9 receiver can both be powered from the BA-48 battery or the later BA-317/U (the common 4 contact connector), this supply also powers the GRC-9 receiver directly as shown above.

It uses the standard CD-1119 GRC-9 Battery Cord as seen here. A future project is to include a +580 VDC GRC-9 transmitter B+ supply inside this box as well. One box will do it all. I just have to find a big DY-88/105 power supply chassis connector.

Look here for further details on that DC-DC power supply:
http://www.n6cc.com/angrc-109-special-forces-radio-set
———————————————————
Problem areas:  Both of my GRC-9’s have been very reliable, they have all the original tubes and capacitors and I have not needed to make any* repairs on them despite 20+ years of frequent use, both mobile and portable. That’s after their military careers.

A “quirk” that the GRC-9 exhibits is a slight chirp when running CW transmissions in VFO mode.  This is due to the design / dynamics of the oscillator running in VFO mode.  Every GRC-9 that I own or have heard exhibits it.  It is quite stable under crystal control however.

It is not objectionable to me or to other operators I have worked.  The other guy knows you are running a classic rig, not some computer with an attached antenna. I call it “character”.

The only item needing attention is the bias battery used to keep the receiver 3Q4 pentode AF power amplifier tube properly biased. (The receiver will work without a battery installed, see below.) I remember recalling that my receivers “sounded” better with the battery installed – and hence the effort to supply that voltage – and the designers thought so as well.

UPDATE: I needed some DATA! The Sylvania tube data for the 3Q4 states that to run the tube as a Class A audio Amplifier, the grid bias needs to be – 4.5 volts and that the audio input not exceed +/- 4.5 volts AC. (It doesn’t want you to drive the grid POSITIVE – the definition of Class A.)

The data sheet states that the tube biased Class A will exhibit 7% distortion. Seems like a lot to me but in keeping the circuit simple and recognizing that its use was not intended to be “Hi Fi”, 7% was deemed to be acceptable.

So I did an experiment. I set up the receiver to hear its crystal calibrator at 2.00 Mc, driving a 600 ohm speaker. (The receiver audio impedance switch was set to 250 ohms for this test and that alone should produce some distortion by itself). This produced a beat note in the receiver while set to CW. I looked at the output with a scope, both with and without the bias battery.

Although I don’t have an audio distortion analyzer, at low audio levels I could see no difference on the scope, the sine wave was very clean in each test. I was a bit surprised. At high audio levels the stage was driven into some distortion and removing the bias made that worse-looking.

I also listened very carefully to bias – no bias condition and I don’t think I can hear any difference at “usual levels” although my hearing has taken a beating over the years. (An Audiophool would probably go apoplectic however…)

Above: Fun on the work bench.

Above: The socket and clamp for the BA-1293 bias battery in an RT-77A. The RT-77 used a hard-wired battery. (The larger diameter + contact is grounded, the -4.5V goes to the smaller socket connector.)

I had previously built replacement bias batteries for that application; I used three 1.5 volt AAAA cells from an ordinary 9 volt Duracell battery. Left in series via the spot-welded stainless steel jumper straps, they make a good replacement battery of -4.5 volts. A bit of heat-shrink tubing or tape, and a salvaged battery connector, it fits in the original battery holder.  That battery will last for years – the shelf life of the cells. Improvise, adapt, overcome.

Note that some of the newer 9 Volt batteries utilize individual rectangular cells, no workie.  This one is fine..

Above: The jury-rigged bias battery as installed in the socket using a defunct BA-1293 battery base and pins (I didn’t have a complete battery with the plastic case). A bit of black electrical tape and heat-shrink tubing holds it together. Works fine but won’t pass inspection….

“If it’s stupid – but works – it isn’t stupid”
Murphy’s Laws of Combat

It may not sound any better in actual practice but the tube will draw its rated plate current and have minimal extra distortion when biased properly into linear, Class A operation. (Why would you operate any tube with a filament, plate, grid or screen voltage widely different from the design voltages and expect good performance?) It’s a big DUH!

Next I looked at the receiver B+ current draw, with and without bias to see if I could detect any change due to the bias. Removing the bias increases the B+ current draw by about 100-200 microamps from an overall 10 milliamps quiescent draw at low output levels. Negligible difference. Probably more of an increase when the volume is turned up – I didn’t test for that.

No surprises here, it’s all about Class A operation. The 3Q4 pentode specs state that it can deliver output with not more than 7% distortion when it is biased correctly for Class A operation. When the Volume is turned up sufficiently to drive the grid positive (with or without the compensating bias) you will get even more distortion. I’m going with the original engineering design – I use the battery. The designer knew what he was doing.
————————————————————————————————————————————————————————————————–
Gremlins Hideaway: One of my GRC-9 receivers had an annoying, intermittent receiver “cut-out” problem that occurred while turning the tuning knob. The receiver would just quit, sometimes preceded by a scratching noise while the knob was being turned.

I assumed it was debris between the tuning capacitor plates causing that (high impedance) stator circuit to short to the grounded rotor plates. I tried several times* to blow the plates clean with a micro duster – that seemed to help – but not permanently – sometimes the problem re-appeared at a different knob position. On the right track.

It started doing it again! I pulled the receiver and removed the shielding covers and blew it out again. Kinda worked. Upon CLOSE examination with a high-magnification eye loupe, I saw a few tiny fibers protruding off the rotor plate but they were not positioned to touch the stator. Ah Ha !! Tin Whiskers.

They were maybe 1-2 mm long, and looked to be about the diameter of spider web silk – if you can get the light just-right to even see them. Maybe a couple of microns in diameter.  If they are here, they are probably elsewhere in the capacitor assembly. They are apparently strong and stiff enough to withstand the micro duster gas blast in some areas that were hard to get to.

Tin whiskers are a well known problem in the electronics business.  They have greatly compromised all kinds of electronics since the environmental-cases started banning lead from tin-lead solder; and apparently even before that. This is the “RoHs” initiative started by the EU and then adopted by the US and others in a knee-jerk reaction.  “A planet completely free of any and all risks.”  But just a little bit of lead alloyed in  significantly inhibits the formation of these tin crystal “whiskers”.

Tin-lead alloy solder does not have this problem; lead-free solder and native tin plating DOES. Some really interesting physics going on here….(If you have done any soldering with the lead-free stuff, you know it does not work very well as a solder, future whiskers notwithstanding).

My radio was made well before that decree started causing short-circuit problems, but the plates of my variable capacitor are apparently tin plated. A thorough “cleaning” with a pipe cleaner or a slice of a business card slid inside each rotor-stator plate gap knocked off any other whiskers that were causing the problem with this set.

Back in operation, no more receiver drop-outs. (By the way, the problem section of C7 was the local oscillator tuning – the LO stopped at certain rotor positions, thus killing the receiver).

Take a look at the tin whiskers on this variable capacitor frame. (Obviously not a GRC-9 capacitor – but you get the idea!)  Holy crap! (Photo: NASA)

According to NASA’s website on Tin Whiskers, at least 10 satellites have been lost and up to 7 nuclear power plants have been partially shut down due to tin whiskers shorting out critical circuits. Is the “cure” worse than the problem? Look at that photo and extrapolate – you be the judge.

—————————————————————————————————————